- Email: [email protected]
Analysis of Team Situation Awareness Using Serious Game and Constructive Model-Based Simulation
13th IFAC/IFIP/IFORS/IEA Symposium on Analysis, Design, and Evaluation of Human-Machine Systems 13th Symposium on Analysis, Design, and Evaluation of Human-Machine Systems 13th IFAC/IFIP/IFORS/IEA IFAC/IFIP/IFORS/IEA Symposium on Aug. 30 - Sept. 2, 2016. Kyoto, Japan 13th IFAC/IFIP/IFORS/IEA Symposium on Analysis, Design, and of Aug. 30 - Sept. 2, 2016. Kyoto, Japan Analysis, Design, and Evaluation Evaluation of Human-Machine Human-Machine Systems Available onlineSystems at www.sciencedirect.com Analysis, Design, and Evaluation of Human-Machine Systems Aug. 2, Kyoto, Aug. 30 30 -- Sept. Sept. 2, 2016. 2016. Kyoto, Japan Japan Aug. 30 - Sept. 2, 2016. Kyoto, Japan
ScienceDirect
Analysis of Team Situation Awareness Using Serious Game and Constructive IFAC-PapersOnLine 49-19 (2016) 537–542 Analysis of Team Situation Awareness Using Serious Game and Constructive Model-Based Simulation Analysis of Team Situation Awareness Using Serious Game and Constructive Analysis of Team Situation Awareness Using Serious Game and Constructive Simulation Analysis of Team SituationModel-Based Awareness Using Serious Game and Constructive Model-Based Simulation Model-Based Simulation Model-Based Simulation Tetsuo Sawaragi*, Kohei Fujii*, Yukio Horiguchi* and Hiroaki Nakanishi* Tetsuo Sawaragi*, Kohei Fujii*, Yukio Horiguchi* and Hiroaki Nakanishi* Tetsuo Sawaragi*, Kohei Fujii*, Fujii*, Yukio Yukio Horiguchi* and Hiroaki Nakanishi* Nakanishi* Tetsuo Sawaragi*, Kohei and Hiroaki * Department of Mechanical andHoriguchi* Science, Kyoto University, Japan. Tetsuo Sawaragi*, Kohei Engineering Fujii*, Yukio and HiroakiKyoto, Nakanishi* * Department of Mechanical Engineering andHoriguchi* Science, Kyoto University, Kyoto, Japan.
: +81-75-383-3581; e-mail: * of Engineering and Science, * Department Department(Tel of Mechanical Mechanical Engineering [email protected]) Science, Kyoto Kyoto University, University, Kyoto, Kyoto, Japan. Japan. : +81-75-383-3581; e-mail: * Department(Tel of Mechanical Engineering [email protected]) Science, Kyoto University, Kyoto, Japan. (Tel : +81-75-383-3581; e-mail: [email protected]) (Tel : +81-75-383-3581; e-mail: [email protected]) (Telanalysis : +81-75-383-3581; e-mail:awareness [email protected]) Abstract: In this study, of team situation using a serious game of supply chain Abstract: In this study, analysis of team situation awareness using a serious game of supply of chain management for steel production is provided. By tracing the development of the performances the Abstract: In this study, analysis of team situation awareness using aa serious game of supply chain Abstract: In this study, analysis of team situation awareness using serious game of supply chain management for steel production is provided. By tracing the development of the performances of the game made by two different teams, a comparative analysis is done with respect to team situation Abstract: In this study, analysis of team situation awareness using a serious game of supply chain management for two steeldifferent production is provided. provided. By tracing tracing the development development ofrespect the performances performances of the the management for steel production is By the of the of game madeshared by teams, a comparative analysis isindividual done withplayers’ to team situation awareness within the team. we into mental models management for two steel production isEspecially, provided. By focus tracing thethe development ofrespect the performances of that the game made by different teams, a comparative analysis is done with to team situation game made by two different teams, a comparative analysis is done with respect to team situation awareness shared within the team. Especially, we focus into the individual players’ mental protocols models that may evolve and grow as the theyteam. accumulate a number ofanalysis game sessions. From the acquired of game made by two different teams, a comparative is done with respect to team situation awareness shared within Especially, we focus into the individual players’ mental models that awareness shared within the team. Especially, we focus into the individual players’ mental models that may evolve and grow as they accumulate a number of game sessions. From the acquired protocols of players during the sessions, players’ temporal mental models of situation awareness are traced, then the awareness shared within the team. Especially, we focus into the individual players’ mental protocols models that may evolve and as accumulate a number of game From the acquired of may evolve andthegrow grow as they they accumulate number of game sessions. From acquired protocols of players during sessions, players’ temporal mental models ofsessions. situation awareness are traced, then the progress of team situation awareness amongaa the players are derived. Through athe comparative analysis of may evolve andthegrow as they accumulate number of game sessions. From the acquired protocols of players during sessions, players’ temporal mental models of situation awareness are traced, then the players during the sessions, players’ temporal mental models of situation awareness are traced, then the progress of team situation awareness among the players are derived. Through a comparative analysis of the two during different teams with respecttemporal to their shared team situation awareness, weare discuss about the players thesituation sessions, players’ mental models of situation awareness traced, then the progress of team awareness among the players are derived. Through a comparative analysis of progress of team situation awareness theshared players are derived. Through a comparative ofa the two different teams with respect to teams their team situation awareness, wethediscuss about the relationships between the with maturity of among the and performance scores attained bywe teams.analysis Finally,the progress of team situation awareness among theshared players are derived. Through a comparative analysis of the two different teams respect to their team situation awareness, discuss about the two different teams with respect to their shared team situation awareness, we discuss about the relationships between the maturity of the teams and performance scores attained by the teams. Finally, a simulation model for generating such multiple players’ behaviors within a variety of teams is proposed. the two different teams with respect to their shared team situation awareness, we discuss about the relationships between the maturity of the teams and performance scores attained by the teams. Finally, aa relationships between the maturity of the teams and performance scores attained by the teams. Finally, simulation model for generating such multiple players’ behaviors within a variety of teams is proposed. This is based uponfor a descriptive decision modelplayers’ of Garbage Can Model, and findings discovered duringa relationships between the maturity of the teams and performance scores byofthe teams. Finally, simulation model generating such multiple behaviors withinattained variety teams is proposed. proposed. simulation model generating such multiple behaviors within aaand variety ofapproach. teams is This is based uponfor aare descriptive decision modelplayers’ of Garbage Can Model, findings discovered during the gaming sessions verified and generalized through a model-based constructive simulation model for generating such multiple players’ behaviors within a variety of teams is proposed. This is upon descriptive decision model of Can and discovered This is based based upon aaare descriptive decision model through of Garbage Garbage Can Model, Model,constructive and findings findings discovered during during the gaming sessions verified and generalized a model-based approach. This is based upon aare descriptive decision model through of Control) Garbage Can Model, and task findings discovered during Keywords: Team situation awareness, collaboration, serious game, cognitive chain the gaming sessions and aa model-based constructive approach. © 2016, IFAC (International Federation of Automatic Hosting by Elsevier Ltd.analysis, All rightssupply reserved. the gaming sessions are verified verified and generalized generalized through model-based constructive approach. Keywords: situation awareness, collaboration, serious game, cognitive task analysis, supply chain the gaming Team sessions arecan verified and generalized through a model-based constructive approach. management, garbage model. Keywords: situation awareness, Keywords: Team Team situation awareness, collaboration, collaboration, serious serious game, game, cognitive cognitive task task analysis, analysis, supply supply chain chain management, garbage can model. Keywords: Team situation awareness, collaboration, serious game, cognitive task analysis, supply chain management, garbage can model. management, garbage can model. management, garbage can model. supply chain, where multiple decision makers in each process 1. INTRODUCTION supply chain, where multiple decision each process of works should collaborate to attain makers higher in productivity as 1. INTRODUCTION supply chain, where multiple decision makers in each supply chain, where multiple decision makers in each process process of works should collaborate to attain higher productivity as 1. well as to recover from the unexpected disturbances. The development and research on situation awareness made supply 1. INTRODUCTION INTRODUCTION chain, where multiple decision makers in each process of works should collaborate to attain higher productivity as of works should collaborate to attain higher productivity as 1. INTRODUCTION well as to recover from the unexpected disturbances. The development and research on situation awareness made Wherein, systematical approaches to higher establishing resilient hugedevelopment advances inand theresearch appliedoncognitive psychologymade and of works should collaborate to attain productivity as well as to recover from the unexpected disturbances. The situation awareness well as to recover from the unexpected disturbances. The development and research on situation awareness made Wherein, systematical approaches to establishing resilient huge advances in the applied cognitive psychology and production and operation systems are needed, and we human factors field (Endsley, 1995). A simplified definition well as to recover from the unexpected disturbances. The development and research on situation awareness made systematical approaches to establishing resilient huge advances in applied cognitive psychology and Wherein, systematical to are establishing resilient huge advances in the the applied psychology and Wherein, production operation systems needed, we human factors field (Endsley, A simplified definition a and means ofapproaches serious games for supplyand chain of situation awareness (SA) is1995). tocognitive know what is happening Wherein, systematical approaches to are establishing resilient huge advances in the applied cognitive psychology and introduce production and operation systems needed, and we human factors field (Endsley, 1995). A simplified definition production and operation systems are needed, and we human factors field (Endsley, 1995). A simplified definition introduce a means of serious games for supply chain of situation awareness (SA) is to know what is happening management for this purpose. around oneself in a complex environment, in order to take an and operation systems are for needed, andchain we human factors field (Endsley, 1995). A simplified definition production introduce a means of serious games supply of situation awareness (SA) is to know what is happening introduce a means of serious games for supply chain of situation awareness (SA) is to know what is happening for this purpose. around oneself in aand complex environment, in order to take an optimal decision, is(SA) an accepted concept ofiscognition in management introduce a means of serious games for supply chain of situation awareness is to know what happening management for this purpose. around oneself in a complex environment, in order to take an games been developed so far, and especially a management forhave this purpose. around oneself in aand complex environment, in order to Though take an optimal decision, is an of cognition in Various complex, socio-technical andaccepted dynamicconcept environments. management forhave this purpose. around oneself in aand complex environment, in order to take an Various games been developed so far, and especially optimal decision, is an accepted concept of cognition in category of serious games is a set of simulations of real-aa optimal decision, and is an accepted concept of cognition in complex, socio-technical and dynamic environments. Though Various games have games been developed developed so far, far, and especially especially the original idea ofand SAisconcerned withconcept the individual decision games have been so and optimal decision, an accepted of cognition in Various category of serious is a set of simulations of real-aa complex, socio-technical and dynamic environments. Though world events or processes designed for the purpose of solving complex, socio-technical and dynamic environments. Though Various games have games been developed so far, and especially the original idea of SA concerned with the individual decision category of serious is a set of simulations of realmakers, it has been extended to an idea of team situation category of serious games is a set of simulations of realcomplex, socio-technical and dynamic environments. Though world events or processes designed serious for the game purpose of solving the original idea of SA concerned the individual decision acategory problem. of thegames well-known is developed the original ideabeen of SAextended concerned with the Salas individual decision ofOne serious is a setfor ofthe simulations of realmakers, it (team has towith an idea of team situation world events or processes designed purpose of solving awareness SA) (Hackerman, 1987; et al., 1995). world events or processes designed for the purpose of solving the original idea of SA concerned with the individual decision aworld problem. One of the infrastructure well-known serious game is developed makers, it has been extended to an idea of team situation Rsolving AIL, to for theevents Dutch railway organization PRO makers, it has been extended to an idea of team situation or processes designed for the purpose of awareness (team SA) (Hackerman, 1987; Salas et al., 1995). a problem. One of the well-known serious game is developed This is it because operators often need and for atest problem. One of innovations the well-known game is developed makers, has been extended to an ideato ofcollaborate team situation RO RAIL, to the Dutch railway organization Poperational awareness (team (Hackerman, 1987; Salas et 1995). new process withserious the railway awareness (team SA) SA) (Hackerman, 1987; Salas et al., al., 1995). problem. One of the infrastructure well-known serious game is developed This is because operators often need to collaborate and afor RO R AIL the Dutch railway infrastructure organization P situation awareness should be extended beyond the individual RO R AIL,, to to for the Dutch railway infrastructure organization P awareness (team SA) (Hackerman, 1987; Salas et al., 1995). test new process innovations with the railway operational This is operators often need to collaborate and chain (Lo et al., 2014). In Japan, the Ironrailway and Steel Institute This is because because operators often need to as collaborate and ROR AIL, to for the Dutch railway infrastructure organization Poperational situation awareness should be extended beyond the individual test new process innovations with the to a team or group level. Team SA is defined "the degree to test new process innovations with the railway operational This is because operators often need to collaborate and chain (Lo et al., 2014). In Japan, the Iron and Steel Institute situation awareness should be extended beyond the (ISIJ) started an academia-industries collaborative situation awareness should be extended the individual individual testJapan new process innovations with the railway operational to a team or group level. Team SA is defined "the degree to of chain (Lo et 2014). In the Iron and Institute which every team member possesses thebeyond SA as required for his chain (Lo et al., al., 2014). In Japan, Japan, the Iron and Steel Steel Institute situation awareness should be extended beyond the individual of Japan (ISIJ) started an academia-industries collaborative to a team or group level. Team SA is defined as "the degree to project from 2012 to 2015 on the topic of human-system coto a team or group level. Team SA is defined as "the degree to chain (Lo et al., 2014). In Japan, the Iron and Steel Institute which every team member possesses the SA required for his of Japan (ISIJ) started an academia-industries collaborative or her responsibilities" (Endsley, 1995, p. 39). The success or of Japanfrom (ISIJ) started an academia-industries collaborative to a team or group level. Team SA is defined as "the degree to creative project 2012 to 2015 on the topic of human-system cowhich every team member possesses the SA required for his risk management for establishing resilience against which every team member possesses the SA required for his of Japan (ISIJ) started an academia-industries collaborative or her responsibilities" (Endsley, 1995, p. 39). The success or project from 2012 to on the topic of cofailure of a team onpossesses the success or failure of each of creative project from 2012 to 2015 2015this on project, the topicMizuyama’s of human-system human-system cowhich every teamdepends member the SA required for his risk management for establishing resilience against or her responsibilities" (Endsley, 1995, p. 39). The success or the disturbances. Within group has or her responsibilities" (Endsley, 1995, p.or 39). The success or project from 2012 to 2015 for on the topic of human-system cofailure of a team depends on the success failure of each of creative risk management establishing resilience against its team members. creative risk management for establishing resilience against or her responsibilities" (Endsley, 1995, p. 39). The success or the disturbances. Mizuyama’s failure of aa team OLagainst PMhas AN developed a Within seriousthis game named Cgroup failure ofmembers. team depends depends on on the the success success or or failure failure of of each each of of the creative risk management forproject, establishing resilience its team disturbances. Within this project, Mizuyama’s group has disturbances. Within this project, Mizuyama’s group has failure ofmembers. a team depends on the success or failure oftheeach of the OLPM AN developed a serious game named C its team (C OLLABORATIVE P RODUCTION M ANAGEMENT ), which is a Current industries are forced to act according to rapid its team members. the disturbances. this project, Mizuyama’s group has PM developed aa Within serious game named C OL PMisAN AN developed serious game named ), which COL its team members. (C OLLABORATIVE P RODUCTION M ANAGEMENT Current industries are forced toas act according to and the torapid virtual supply chain of a large-scale make-to-order company changes of circumstances such market changes the developed a Pserious game named ), which COLPMisANaa (C OLLABORATIVE RODUCTION M ANAGEMENT Current industries are forced to act according to the rapid (C OLLABORATIVE P RODUCTION M ANAGEMENT ), which is eta Current industries are forced to act according to the rapid virtual supply chain of a large-scale make-to-order company changes of circumstances such as market changes and to the through makingchain production and delivery decisions (Nonaka change of economicareenvironments. Moreover, the reduction (C OLLABORATIVE PRODUCTION MANAGEMENT ), which is a Current industries forced toas act according to and the torapid virtual supply of aa large-scale make-to-order company changes of circumstances such market changes the virtual supply chain of large-scale make-to-order company changes of circumstances such as market changes and to the through making production and delivery decisions (Nonaka et change of economic environments. Moreover, the reduction al., 2016). Thechain general of this game is twofold; to of expertof workers may bring about drop ofchanges the the work quality, virtual supply of apurpose large-scale make-to-order company changes of circumstances such as market and to the through making production and delivery decisions (Nonaka et change economic environments. Moreover, reduction through making production and delivery decisions (Nonaka et change of economic environments. Moreover, the reduction al., 2016). The general purpose of this game is twofold; to of expert workers may bring about drop of the work quality, games as a platform forand policy development, such as the and theofoccurrences ofbring malfunctions caused by reduction aging of use through making production delivery decisions (Nonaka et change economic environments. Moreover, the al., 2016). The general purpose of this game is twofold; to of expert workers may about drop of the work quality, al., 2016). The general purpose of this game is twofold; to of expert workers may bring about drop of the work quality, use games as a platform for policy development, such as the and the occurrences of malfunctions caused by aging of introduction ofa platform new management forisattaining the equipment may may bring about a drop variety of work unexpected al., 2016). as The general purpose ofconcepts this game twofold; to of expert workers bring about of the quality, use for policy development, such as and the of caused aging of use games games asof a platform for policyconcepts development, such as the the and the occurrences occurrences of malfunctions malfunctions caused byunexpected aging of resilient introduction new management for attaining equipment may bring about a the variety of by and coordinated production scheduling by multiple fluctuations frequently during production. Taking use games as a platform for policy development, such as the and the occurrences of malfunctions caused by aging of introduction new management concepts for attaining the equipment may bring about aa the variety of unexpected introduction of new management concepts forsecondly attaining thea equipment may variety of through unexpected and of coordinated scheduling by multiple fluctuations frequently during production. Taking actors consisting of management the production supply chain, andfor asthe measures suited tobring those about occasions is realized the resilient introduction of new concepts attaining equipment may bring about a the variety of unexpected resilient and coordinated production scheduling by multiple fluctuations frequently during production. Taking resilient and coordinated production scheduling by multiple fluctuations frequently during the production. Taking of hypotheses the production supply on chain, andbehaviors secondly as a measures suited to those during occasions realized through the research tool coordinated to test human within human-system co-creative safety management for actors resilientconsisting and scheduling by multiple fluctuations frequently theis production. Taking actors consisting of the chain, and secondly as measures suited to those occasions is realized through the actors consisting of hypotheses the supply supply chain, andbehaviors secondlywithin as aa measures suited to those occasions is of realized throughThis the research tool to test on human human-system co-creative safety management for the supply chain management system. establishing resilience against a variety disturbances. actors consisting of hypotheses the supply on chain, andbehaviors secondlywithin as a measures suited to those occasions is realized through the research tool to test human human-system co-creative safety management for research tool to management test hypotheses on human behaviors within human-system safety of management for the supplytool chain system. establishing a variety disturbances. This situation is resilience also co-creative true against and becoming serious in the steel research towe testuse hypotheses on human within human-system co-creative safety of management for In the supply chain management system. establishing resilience against aa variety disturbances. This PMAN game to this paper, this testbed of COLbehaviors the supply chain management system. establishing resilience against variety of disturbances. This situation is also true and becoming serious in the steel production industries. Steel production of a the complex the supply chain management system. establishing resilience against a variety consists of disturbances. This In OLPM AN game this paper, we use this testbed of C situation is also true and becoming serious in steel analyze team situation awareness developed during a series situation isindustries. also trueSteel and production becoming consists serious of in a the steel In this paper, we use this testbed of COLPMAN game to production complex to AN game to this paper, we use this testbeddeveloped of COLPM situation isindustries. also trueSteel and production becoming consists serious of in a the steel In analyze team situation during a series production complex production industries. Steel production consists of a complex In AN game to this paper, we use awareness this testbeddeveloped of COLPM analyze team situation awareness during a series analyze team situation awareness developed during a series production industries. Steel production consists of a complex Copyright © 2016 IFAC 548 analyze team situation awareness developed during a series Copyright © 2016, 2016 IFAC 548 Hosting by Elsevier Ltd. All rights reserved. 2405-8963 © IFAC (International Federation of Automatic Control) Copyright © 2016 IFAC 548 Copyright 2016 responsibility IFAC 548Control. Peer review©under of International Federation of Automatic Copyright © 2016 IFAC 548 10.1016/j.ifacol.2016.10.617
2016 IFAC/IFIP/IFORS/IEA HMS 538 Aug. 30 - Sept. 2, 2016. Kyoto, Japan
Tetsuo Sawaragi et al. / IFAC-PapersOnLine 49-19 (2016) 537–542
of game sessions performed by a couple of teams, each of which consists of multiple players. Based on the analysis of players’ decisions and interaction behaviours, some key issues for team situation awareness are identified, and then a novel constructive approach to replaying behaviours of the players are introduced. That is, a simulation model for generating such multiple players’ behaviours within a variety of teams is proposed. This is based upon a descriptive decision model of Garbage Can Model (Cohen et al., 1972), and findings discovered during the empirical analysis of gaming sessions are verified and generalized through a model-based constructive approach.
2. OVERVIEW OF THE GAME COLPMAN is a serious game of supply chain management for steel production. Players of the game consists of the five players; a headquarter accepting orders from customers, an upstream factory producing materials, and three downstream factories processing materials into products that are to be delivered to the customers. This game operates a virtual supply chain of a large-scale make-to-order company through making production and delivery decisions. Details of the game are provided in Nonaka et al. (2016) of this session.
When the schedules are fixed, productions in each factory start and materials produced in a upstream factory are delivered to down stream factories, while downstream factories continue their productions using the supplied materials producing the final products delivered to customers. Wherein, a variety of costs are defined; costs for stocks, costs for delivery, costs for set-up change, and penalties for late delivery. Five players should collaborate to make their decisions on making production schedules and ordering by communicating with each other to maximize sales volumes and minimizing costs and penalties. The game platform consists of decision-making sessions and a simulating part that simulates productions in factories and flows of goods resulted by the former decisions. In the simulation part, a variety of fluctuations such as the ones of lead times, occurrences of manufacturing defects and arrivals of express orders are embedded stochastically. Therefore, each player has to carefully adjust and modify their ongoing plans by communicating the other players in necessity. The game session consists of a series of trials, where each trial is composed of four terms and each term is divided into four periods. At the beginning period of each term, regular orders from customers come to headquarter and all five players start scheduling the production by making material orders in discussing each other, and when they are fixed they are sent to the simulating part that simulate production and deliveries period by period. Because of accidental occurrences of fluctuations, each player has to adjust their decisions at the beginning of the subsequent period in monitoring the production status.
3. COGNITIVE TASK ANALYSIS OF THE GAME
Fig.1 The flow chart of the COLPMAN game Figure 1 illustrates flows of goods and information among the players. Game starts when headquarter accepts orders from customers. An order is defined as a triple consisting of customer, type (of five types), and size (of five sizes) of goods to order. Customers are assumed to be located in places with different distances from downstream factories which supply products to the customers, i.e., times needed to deliver products to customers are different due to a distance gap. Then, headquarter makes a rough plan by assigning downstream factories to produce ordered goods as well as determining appointed dates of delivery of each order. Each of assigned downstream factories makes its own schedule of producing the assigned products as well as making orders of the materials needed for that production to an upstream factory. Orders from the downstream factories to the upstream factory are specified as a tuple of quantity and types of material. Then, accepting orders from downstream factories upstream factory makes its schedule to produce materials to be supplied to the downstream factories. Both of upstream and downstream factories have some amounts of stocks of materials and products, respectively, and should manage these not to produce excess stocks. 549
Despite the large volume of researches on situation awareness across domains, there proposed numerous definitions of SA. One of the most accepted definitions of situation awareness is that of Endsley (Endsley, 1995), called three-level model. Endsley extended a single person’s SA into team SA, where teams are recognized as unit of analysis and team SA is more than a simple sum of individual SA. Related to this, Endsley also defined shared situation awareness as "the degree to which team members possess the same SA on shared SA requirements" (Endsley & Jones, 1997). There are information requirements that are relevant to multiple team members, and a major part of teamwork involves the area where these SA requirements overlap. During COLPMAN game sessions performed by five human players, each player has a subgoal pertinent to his/her specific role that feeds into the overall team goal. Associated with each member's subgoal a set of SA elements about which he/she is concerned exists. As the members of a team are essentially interdependent in meeting the overall team goal, some overlap between each member's subgoal and their SA requirements will be present. To measure and visualize team SA acquired during the game sessions, we introduce a function hierarchy for the game, in which a set of SA elements to be shared with others are structurally configured with different abstraction levels to analyze the shared
2016 IFAC/IFIP/IFORS/IEA HMS Aug. 30 - Sept. 2, 2016. Kyoto, Japan
Tetsuo Sawaragi et al. / IFAC-PapersOnLine 49-19 (2016) 537–542
Fig.2 A functional hierarchy for the the COLPMAN game
539
and incurred in the trial, respectively. Individual trials were tried in one day with no time limit, and subsequent trials were done on different days by keeping regular time intervals. Assignments of the roles within the teams were made at random initially and were kept fixed during the trials till the end of the games. The scores of the game for the teams were recorded for each trial, and the game was reset whenever the teams started new trials. Before starting the game sessions each team was allowed to perform the game as many as three trials to understand the gaming procedures, when the collaborative learning among the players was inhibited. In the gaming sessions, participants are allowed to communicate with other teammates as they like, and logging data of the conversation made during the sessions were all recorded using multiple video cameras. The recorded conversations of the two teams were analysed for each trial. Utterances of the individual players were traced each by each, and they were mapped onto the appropriate nodes within their individual function hierarchies. Thus, a series of function hierarchies in which some parts of the hierarchy were marked highlighted representing temporal single SA of individual players were obtained at the ends of every trial. Then, evolutions of single SA for every player as well as maturity of team SA of the teams were analysed for comparison.
elements of SA. This is illustrated in Fig.2. A function hierarchy is organized in three major abstraction layers, where a top-most layer represents goals to be attained for a successful game play and a second layer consists of the subgoals to attain the goals. Bottom layer represents a set of operations to realize the respective subgoals in the second layer. This hierarchy is defining relationships among the SA elements connected by means-ends relations that are to be shared among the players of the team. In this work, we use this function hierarchy as a common template for specifying the individual SA of each player as well as for evaluating shared SA among the players. The conversations, i.e., protocols, made during the game sessions as well as the game logging data showing the players’ decisions and status of the productions are all recorded. From these data, it is identified which parts of the function hierarchy an individual player is aware of in each stages of the game sessions. Therefore, by highlighting the parts of the hierarchy in awareness, each player’s cognition vector can be defined as his/her individual SA. Moreover, different players’ cognition vectors can be used to evaluate similarities and overlaps of their cognitions, thus shared SA can be quantified dynamically as the game sessions proceeds.
4.2 Results of Experiments A. Evolutions of Team SA: Throughness Attained by Teams Figure 3 shows the scores of the game obtained by team A and team B as the game sessions proceeds as many as six trials; connected line plots show the score obtained and vertical bar charts represents the number of items, i.e., nodes of function hierarchies, at the levels of subgoals and subsubgoals that are shared by more than two players within the team at the end of trials. Each bar chart is divided into different colored parts according to the abstraction levels of a hierarchy, where numbers shows items shared by players.
4. EXPERIMENTS OF GAME 4.1 Overview of Experiments At first we made a series of experiments using the game COLPMAN and investigated into how team SA evolves as the team consisting of five human players iterates and accumulates the gaming sessions. For this purpose, ten male postgraduate school students participated in the experiments, who have no expertise on supply chain management of steel productions. These ten participants were divided into two teams, team A and team B.
As for the scores attained by the teams, team B is achieving a better result early in the trials as compared with team A, while team A achieves worse in the beginning and then gradually improves the score and finally reached to the same level of the score with team B. The number of items shared within the teams and their evolutions are quite different between the two teams; team B has much shared SA among the players even from the early stages of the trials, while
Each team was requested to perform the game sessions with the same teammates for six trials, where a single trial consists of four terms. Participants were requested to gain the score as higher as possible, where the score was calculated from sales volumes and cost/penalties attained 550
2016 IFAC/IFIP/IFORS/IEA HMS 540 Aug. 30 - Sept. 2, 2016. Kyoto, Japan
Tetsuo Sawaragi et al. / IFAC-PapersOnLine 49-19 (2016) 537–542
team A needs more trials to get to share SA within the team. The similar results were obtained as for the comparison with respect to the items corresponding to the bottom layer in the hierarchies.
The results show that in team A the lengths of the fragments are increasing as the trials go on, but the numbers of players contributing to that problem solving are less as compared with team B. On the other hand, in team B conversations lasted longer and more players are contributing that problem solving. 4.3 Findings Obtained from Experiments From the results of the experiments mentioned in the previous section, aspects of team SA for the two teams are quite different and are characterized as follows. Team A: The averaged efforts contributed by the players are less and the team is apt to solve problems on the items that are located in lower portions of the function hierarchy preferring solving operational items to sharing more fundamental issues that are located in the upper parts of the hierarchy.
Fig.3 Results of experiments
B. Evolutions of Performance: Efficiency of Problem Solving by Teams
Team B: The averaged efforts contributed by the players are significant with respect to the lengths of the conversation fragments as well as to the number of players who contributed to collaboration. The team prefers sharing items located in the middle of the function hierarchy to the operational ones in the bottom.
More detailed analysis on the protocols obtained during the game sessions was made to see why the resulted team SA was different between the teams. For this purpose, a chronological series of utterances were segmented into fragments in which utterances were made on the continuous related topics. By tracing the chronological order of the utterances by the players and checking the contents of those, if the transitions of the utterances were on the connected parts of the items within the hierarchies, those were regarded as parts contributing to problem solving on the common issues, while the transitions among the apart portions of the hierarchies are regarded as turnovers of the problems that the team attempted to solve. By tracing the conversations starting from the triggering utterances and segmenting these into a set of fragments according to the above rules, the performances of the two teams were characterized.
5. CONSTRUCTIVE APPROACH TO TEAM SA USING GARBAGE CAN MODEL 5.1 Cohen’s Garbage Can Model The findings mentioned in the previous section were results specific to the two teams who participated in the experiments. Then, we attempt to generalize the empirical findings obtained so far so that we could gain more generalized insights as for the team SA and for designing good teams with respect to SA. For this purpose, we take a constructive approach. The constructive approach is a scientific methodology in which an objective system is to be understood by constructing the system and operating it. We can identify two types of methodologies in simulation studies, one is realistic simulation and the other is constructive simulation. The former tries to make operational copies of actual phenomena as possible as realistic in order to predict what occur in the target phenomena. The latter constructs rather simplified non-realistic models in order to extract essential features and to understand underlying mechanisms and logics of the objective phenomena. In this work, we present a simulation testbed by which human dynamic interactions within the team members during COLPMAN game sessions based on the second simulation concept.
Fig.4 Evolution of utterances during each trial Figure 4 illustrates the results of the number of utterances that were exchanged among the players for solving common problems collaboratively, i.e., lengths of the fragments contributing to the same issues. This graph also illustrates a number of players who joined to collaborative problem solving, i.e., a number of players who participated in the common problem solving in their utterances.
For this purpose, we introduce Cohen’s Garbage Can Model (GCM) that was originally introduced in a field of organization theory for explaining anarchy of organizational behaviors. We extend the original model for finding out the bottlenecks in which the Efficiency and Throughness TradeOff (ETTO) principle of the team is preserved and/or broken by varying a set of essential parameters determining the team characteristics (Hollnagel, 2009). 551
2016 IFAC/IFIP/IFORS/IEA HMS Aug. 30 - Sept. 2, 2016. Kyoto, Japan
Tetsuo Sawaragi et al. / IFAC-PapersOnLine 49-19 (2016) 537–542
The principal components of the GCM are as follows: problems that are concerns of people inside the team, choice opportunities that are occasions when a team is expected to produce behavior that can be called a decision, and participants as human resources of the team contributing to decision-making.
541
evolve according to which access strategy is adopted. ppotential = 1.0 means that a team always takes an access strategy of decision structure A, while ppotential = 0.0 means that a team always takes an access strategy of decision structure B. A decision structure A and B are set corresponding to the findings observed during the game session experiments on team A and B, respectively, mentioned in section 4.
Participants as well as problems are thrown into choice opportunities that are modeled as “garbage cans”, and when the total amount of energy thrown into the garbage can exceeds the energy that is required to solve the problems thrown into that garbage can, that choice opportunity is regarded as resolved, and then they disappear from the organization. The constraints on which problem and which participant is thrown into which choice opportunities can be varied (i.e., an access structure and a decision structure, respectively). Under these constraints, problems as well as participants are allowed to be allocated into one of the choice opportunities, but even after they are thrown into to a particular choice opportunity, they are allowed to move to another choice opportunity anytime whenever some criteria is satisfied. That is, problems may disappear from that choice opportunity and participants may come and go. In this way, the garbage can model can simulate a descriptive organizational behavior of problem solving.
5.3 Results of Team SA Simulations Using GCM
5.2 Mapping COLPMAN game into GCM In our modeling, problems of GCM correspond to the items making up the functional hierarchy of COLPMAN game, and we regard individual team players as participants of GCM. Choice opportunities of GCM correspond to a set of fragments segmented from a series of conversations made during the game sessions. An access structure, i.e., a constraint between choice opportunities and assignable problems, is defined according to the correspondences between fragments of the conversation and pre-classified groups of items in the functional hierarchy that are referred to during the conversation.
Fig.5 Results of GCM simulations Figure 5(a) shows how distributions of different types of problem solving, i.e., decision by resolution, decision by oversight, and decision by flight, vary according to a parameter of ppotential. A type of decision by resolution shows a choice that is made at an opportunity after all contained problems are resolved, and a type of decision by flight shows a choice that is made at an opportunity in spite of some problems not yet made resolved. A type of decision by oversight shows a choice opportunity is set up by some players but no one commit to that then disappears from the subsequent conversation for a while. Figure 5(b) shows how much efforts of the players are needed to reach to problem solving by resolution on average, where efforts are indexed as a number of steps of players’ commitments. Other results showing how the team SA evolves concerning with a number of players who shared the cognition are obtained as well (i.e., derived from players’ cognitive vectors). All the simulation results are averaged results gained from simulations iterated as many as 1000 times.
In addition to a set of original parameters we add a novel parameter of each player’s cognition vector showing his/her temporal awareness on the items of the hierarchy, which is stochastically updated when game players commit to a choice opportunity and the items of that choice opportunity are regarded as cognized by that player. As for a constraint of decision structure, i.e., a constraint between choice opportunities and their committable players, we define two variations as follows. Decision Structure A: This constraint represents that players are apt to selectively commit to the choice opportunity where problem solving by resolution is mostly expected. Decision Structure B: This constraint represents that players are apt to selectively commit to the choice opportunity where the more players are already committed.
5.4 Towards Designing Good Teams
We define a probabilistic parameter ppotential meaning a probability that a team selectively adopts their access strategies according to the decision structure A or B, and by varying this parameter we simulate how the team behaviors
In addition to replaying the observed behaviors of our empirical experiments, we investigate into how our simulation model is used for the purpose of team design. There would be so many factors that can affect the team 552
2016 IFAC/IFIP/IFORS/IEA HMS 542 Aug. 30 - Sept. 2, 2016. Kyoto, Japan
Tetsuo Sawaragi et al. / IFAC-PapersOnLine 49-19 (2016) 537–542
performances, some of which are identified as a number of parameters defined in the Garbage Can Model, such as selection of two constraints, selection of participants, distribution of the expertise among the participants, etc. Team performance should be evaluated with respect to their efficiencies, thoroughness, depth and broadness of team SA, etc.
division of roles within the team, homogeneity and/or heterogeneity of tam members with respect to their expertise, etc., but our proposal of constructive approach does contribute to discovering a novel way for team fostering and for human resource management.
6. CONCLUSIONS In this work, analysis of team situation awareness using serious game of supply chain management for steel production was provided. Especially, we focus into the individual players’ mental models and we introduced a function hierarchy tree to visualize players’ temporal mental models through cognitive analysis. Through a comparative analysis of the two different teams with respect to their shared team situation awareness, we discussed about the relationships between the maturity of the shared mental models and performance scores attained by the teams. Then, a simulation model for generating such multiple players’ behaviours within a variety of teams was proposed. We showed findings discovered during the gaming sessions are verified and generalized through a model-based constructive approach, and discussed about its extension towards developing a novel approach to team design.
Fig.6 Simulated results when external burdens are increased for a virtual team having decision structure A In spite of many freedoms for team design, the essence is that a team behaviour is not the single sum of individual decision-makers or problem solvers, but is rather an “open” and “ambiguous” process, and the management for such a complex system will be a critical issue especially when the team faces with unexpected disturbances. Human dynamic interactions within the team are determined not only specifications of the team per se but also by what are externally provided to the team and how the team adapts to such external disturbances.
REFERENCES Cohen, M.D., March, J.G. and Olsen, J.P. (1972). A Garbage Can Model of Organizational Choice, Administrative Science Quarterly, 17(1). Endsley, M.R. (1995). Toward a Theory of Situation Awareness in Dynamic Systems, Human Factors, 37(1), pp.32–64. Endsley, M. & Jones, W. (1997). Situation awareness, information dominance, and information warfare, Wright-Patterson AFB, OH: United States Air Force Armstrong Laboratory. Hackman,J.R. (1987). The design of work teams, Handbook of organizational behavior, pp.315-342. Hollnagel, E. (2009). The ETTO principle: efficiencythoroughness trade-off: why things that go right sometimes go wrong. Aldershot, Hants, England: Ashgate. LoJ.C., van den Hoogen, J. Meijer, S.A. (2014). Testing Changes in Railway System Through Gaming Simulation: How Different Types of Innovations Affect Operators’ Mental Models Proceedings of the 5th International Conference on Applied Human Factors and Economics AHFE 2014. Nonaka, T, Miki, K., Odajima, J. and Mizuyama, H. (2016). Analysis of Dynamic Decision Making Underpinning Supply Chain Resilience: A Serious Game Approach, to be presented at IFAC-HMS 2016. Salas, E., Prince, C., Baker, D.P., Shrestha, L. (1995). Situation Awareness in Team Performance: Implications for Measurement and Training, Human Factors, 37 (1), pp.123-136.
In this work, using the developed simulation system for COLPMAN game we investigated into the relation between team behaviours and external workloads that are thrown into the game. If we were to discover relations between those, we could find out an indirect way for designing the team by controlling the external workloads. This is a so-called indirect control in contrast to the conventional direct control. The indirect control method guides the system by controlling the system's boundary conditions and is a concept in opposition to the direct control method that controls the parameters in the system. By applying this method to the virtual team, we expect to realize an effective method of guiding the team to evolve into a mature team. Figure 6 shows the results of the simulation when external burdens are varied for a team characterized with the feature of decision structure A. The results show that problem solving efficiencies are degraded due to the difficulties of problem solving thrown into the team. However, more minute investigation into how their team SA evolves, the result shows both the depth and the broadness of team SA are improved more as compared with team SA cultivated for less burdens. This shows a possible way of nurturing teams by controlling the provision of external workloads. The result may indeed depend upon the team characteristics,
553
ScienceDirect
Analysis of Team Situation Awareness Using Serious Game and Constructive IFAC-PapersOnLine 49-19 (2016) 537–542 Analysis of Team Situation Awareness Using Serious Game and Constructive Model-Based Simulation Analysis of Team Situation Awareness Using Serious Game and Constructive Analysis of Team Situation Awareness Using Serious Game and Constructive Simulation Analysis of Team SituationModel-Based Awareness Using Serious Game and Constructive Model-Based Simulation Model-Based Simulation Model-Based Simulation Tetsuo Sawaragi*, Kohei Fujii*, Yukio Horiguchi* and Hiroaki Nakanishi* Tetsuo Sawaragi*, Kohei Fujii*, Yukio Horiguchi* and Hiroaki Nakanishi* Tetsuo Sawaragi*, Kohei Fujii*, Fujii*, Yukio Yukio Horiguchi* and Hiroaki Nakanishi* Nakanishi* Tetsuo Sawaragi*, Kohei and Hiroaki * Department of Mechanical andHoriguchi* Science, Kyoto University, Japan. Tetsuo Sawaragi*, Kohei Engineering Fujii*, Yukio and HiroakiKyoto, Nakanishi* * Department of Mechanical Engineering andHoriguchi* Science, Kyoto University, Kyoto, Japan.
: +81-75-383-3581; e-mail: * of Engineering and Science, * Department Department(Tel of Mechanical Mechanical Engineering [email protected]) Science, Kyoto Kyoto University, University, Kyoto, Kyoto, Japan. Japan. : +81-75-383-3581; e-mail: * Department(Tel of Mechanical Engineering [email protected]) Science, Kyoto University, Kyoto, Japan. (Tel : +81-75-383-3581; e-mail: [email protected]) (Tel : +81-75-383-3581; e-mail: [email protected]) (Telanalysis : +81-75-383-3581; e-mail:awareness [email protected]) Abstract: In this study, of team situation using a serious game of supply chain Abstract: In this study, analysis of team situation awareness using a serious game of supply of chain management for steel production is provided. By tracing the development of the performances the Abstract: In this study, analysis of team situation awareness using aa serious game of supply chain Abstract: In this study, analysis of team situation awareness using serious game of supply chain management for steel production is provided. By tracing the development of the performances of the game made by two different teams, a comparative analysis is done with respect to team situation Abstract: In this study, analysis of team situation awareness using a serious game of supply chain management for two steeldifferent production is provided. provided. By tracing tracing the development development ofrespect the performances performances of the the management for steel production is By the of the of game madeshared by teams, a comparative analysis isindividual done withplayers’ to team situation awareness within the team. we into mental models management for two steel production isEspecially, provided. By focus tracing thethe development ofrespect the performances of that the game made by different teams, a comparative analysis is done with to team situation game made by two different teams, a comparative analysis is done with respect to team situation awareness shared within the team. Especially, we focus into the individual players’ mental protocols models that may evolve and grow as the theyteam. accumulate a number ofanalysis game sessions. From the acquired of game made by two different teams, a comparative is done with respect to team situation awareness shared within Especially, we focus into the individual players’ mental models that awareness shared within the team. Especially, we focus into the individual players’ mental models that may evolve and grow as they accumulate a number of game sessions. From the acquired protocols of players during the sessions, players’ temporal mental models of situation awareness are traced, then the awareness shared within the team. Especially, we focus into the individual players’ mental protocols models that may evolve and as accumulate a number of game From the acquired of may evolve andthegrow grow as they they accumulate number of game sessions. From acquired protocols of players during sessions, players’ temporal mental models ofsessions. situation awareness are traced, then the progress of team situation awareness amongaa the players are derived. Through athe comparative analysis of may evolve andthegrow as they accumulate number of game sessions. From the acquired protocols of players during sessions, players’ temporal mental models of situation awareness are traced, then the players during the sessions, players’ temporal mental models of situation awareness are traced, then the progress of team situation awareness among the players are derived. Through a comparative analysis of the two during different teams with respecttemporal to their shared team situation awareness, weare discuss about the players thesituation sessions, players’ mental models of situation awareness traced, then the progress of team awareness among the players are derived. Through a comparative analysis of progress of team situation awareness theshared players are derived. Through a comparative ofa the two different teams with respect to teams their team situation awareness, wethediscuss about the relationships between the with maturity of among the and performance scores attained bywe teams.analysis Finally,the progress of team situation awareness among theshared players are derived. Through a comparative analysis of the two different teams respect to their team situation awareness, discuss about the two different teams with respect to their shared team situation awareness, we discuss about the relationships between the maturity of the teams and performance scores attained by the teams. Finally, a simulation model for generating such multiple players’ behaviors within a variety of teams is proposed. the two different teams with respect to their shared team situation awareness, we discuss about the relationships between the maturity of the teams and performance scores attained by the teams. Finally, aa relationships between the maturity of the teams and performance scores attained by the teams. Finally, simulation model for generating such multiple players’ behaviors within a variety of teams is proposed. This is based uponfor a descriptive decision modelplayers’ of Garbage Can Model, and findings discovered duringa relationships between the maturity of the teams and performance scores byofthe teams. Finally, simulation model generating such multiple behaviors withinattained variety teams is proposed. proposed. simulation model generating such multiple behaviors within aaand variety ofapproach. teams is This is based uponfor aare descriptive decision modelplayers’ of Garbage Can Model, findings discovered during the gaming sessions verified and generalized through a model-based constructive simulation model for generating such multiple players’ behaviors within a variety of teams is proposed. This is upon descriptive decision model of Can and discovered This is based based upon aaare descriptive decision model through of Garbage Garbage Can Model, Model,constructive and findings findings discovered during during the gaming sessions verified and generalized a model-based approach. This is based upon aare descriptive decision model through of Control) Garbage Can Model, and task findings discovered during Keywords: Team situation awareness, collaboration, serious game, cognitive chain the gaming sessions and aa model-based constructive approach. © 2016, IFAC (International Federation of Automatic Hosting by Elsevier Ltd.analysis, All rightssupply reserved. the gaming sessions are verified verified and generalized generalized through model-based constructive approach. Keywords: situation awareness, collaboration, serious game, cognitive task analysis, supply chain the gaming Team sessions arecan verified and generalized through a model-based constructive approach. management, garbage model. Keywords: situation awareness, Keywords: Team Team situation awareness, collaboration, collaboration, serious serious game, game, cognitive cognitive task task analysis, analysis, supply supply chain chain management, garbage can model. Keywords: Team situation awareness, collaboration, serious game, cognitive task analysis, supply chain management, garbage can model. management, garbage can model. management, garbage can model. supply chain, where multiple decision makers in each process 1. INTRODUCTION supply chain, where multiple decision each process of works should collaborate to attain makers higher in productivity as 1. INTRODUCTION supply chain, where multiple decision makers in each supply chain, where multiple decision makers in each process process of works should collaborate to attain higher productivity as 1. well as to recover from the unexpected disturbances. The development and research on situation awareness made supply 1. INTRODUCTION INTRODUCTION chain, where multiple decision makers in each process of works should collaborate to attain higher productivity as of works should collaborate to attain higher productivity as 1. INTRODUCTION well as to recover from the unexpected disturbances. The development and research on situation awareness made Wherein, systematical approaches to higher establishing resilient hugedevelopment advances inand theresearch appliedoncognitive psychologymade and of works should collaborate to attain productivity as well as to recover from the unexpected disturbances. The situation awareness well as to recover from the unexpected disturbances. The development and research on situation awareness made Wherein, systematical approaches to establishing resilient huge advances in the applied cognitive psychology and production and operation systems are needed, and we human factors field (Endsley, 1995). A simplified definition well as to recover from the unexpected disturbances. The development and research on situation awareness made systematical approaches to establishing resilient huge advances in applied cognitive psychology and Wherein, systematical to are establishing resilient huge advances in the the applied psychology and Wherein, production operation systems needed, we human factors field (Endsley, A simplified definition a and means ofapproaches serious games for supplyand chain of situation awareness (SA) is1995). tocognitive know what is happening Wherein, systematical approaches to are establishing resilient huge advances in the applied cognitive psychology and introduce production and operation systems needed, and we human factors field (Endsley, 1995). A simplified definition production and operation systems are needed, and we human factors field (Endsley, 1995). A simplified definition introduce a means of serious games for supply chain of situation awareness (SA) is to know what is happening management for this purpose. around oneself in a complex environment, in order to take an and operation systems are for needed, andchain we human factors field (Endsley, 1995). A simplified definition production introduce a means of serious games supply of situation awareness (SA) is to know what is happening introduce a means of serious games for supply chain of situation awareness (SA) is to know what is happening for this purpose. around oneself in aand complex environment, in order to take an optimal decision, is(SA) an accepted concept ofiscognition in management introduce a means of serious games for supply chain of situation awareness is to know what happening management for this purpose. around oneself in a complex environment, in order to take an games been developed so far, and especially a management forhave this purpose. around oneself in aand complex environment, in order to Though take an optimal decision, is an of cognition in Various complex, socio-technical andaccepted dynamicconcept environments. management forhave this purpose. around oneself in aand complex environment, in order to take an Various games been developed so far, and especially optimal decision, is an accepted concept of cognition in category of serious games is a set of simulations of real-aa optimal decision, and is an accepted concept of cognition in complex, socio-technical and dynamic environments. Though Various games have games been developed developed so far, far, and especially especially the original idea ofand SAisconcerned withconcept the individual decision games have been so and optimal decision, an accepted of cognition in Various category of serious is a set of simulations of real-aa complex, socio-technical and dynamic environments. Though world events or processes designed for the purpose of solving complex, socio-technical and dynamic environments. Though Various games have games been developed so far, and especially the original idea of SA concerned with the individual decision category of serious is a set of simulations of realmakers, it has been extended to an idea of team situation category of serious games is a set of simulations of realcomplex, socio-technical and dynamic environments. Though world events or processes designed serious for the game purpose of solving the original idea of SA concerned the individual decision acategory problem. of thegames well-known is developed the original ideabeen of SAextended concerned with the Salas individual decision ofOne serious is a setfor ofthe simulations of realmakers, it (team has towith an idea of team situation world events or processes designed purpose of solving awareness SA) (Hackerman, 1987; et al., 1995). world events or processes designed for the purpose of solving the original idea of SA concerned with the individual decision aworld problem. One of the infrastructure well-known serious game is developed makers, it has been extended to an idea of team situation Rsolving AIL, to for theevents Dutch railway organization PRO makers, it has been extended to an idea of team situation or processes designed for the purpose of awareness (team SA) (Hackerman, 1987; Salas et al., 1995). a problem. One of the well-known serious game is developed This is it because operators often need and for atest problem. One of innovations the well-known game is developed makers, has been extended to an ideato ofcollaborate team situation RO RAIL, to the Dutch railway organization Poperational awareness (team (Hackerman, 1987; Salas et 1995). new process withserious the railway awareness (team SA) SA) (Hackerman, 1987; Salas et al., al., 1995). problem. One of the infrastructure well-known serious game is developed This is because operators often need to collaborate and afor RO R AIL the Dutch railway infrastructure organization P situation awareness should be extended beyond the individual RO R AIL,, to to for the Dutch railway infrastructure organization P awareness (team SA) (Hackerman, 1987; Salas et al., 1995). test new process innovations with the railway operational This is operators often need to collaborate and chain (Lo et al., 2014). In Japan, the Ironrailway and Steel Institute This is because because operators often need to as collaborate and ROR AIL, to for the Dutch railway infrastructure organization Poperational situation awareness should be extended beyond the individual test new process innovations with the to a team or group level. Team SA is defined "the degree to test new process innovations with the railway operational This is because operators often need to collaborate and chain (Lo et al., 2014). In Japan, the Iron and Steel Institute situation awareness should be extended beyond the (ISIJ) started an academia-industries collaborative situation awareness should be extended the individual individual testJapan new process innovations with the railway operational to a team or group level. Team SA is defined "the degree to of chain (Lo et 2014). In the Iron and Institute which every team member possesses thebeyond SA as required for his chain (Lo et al., al., 2014). In Japan, Japan, the Iron and Steel Steel Institute situation awareness should be extended beyond the individual of Japan (ISIJ) started an academia-industries collaborative to a team or group level. Team SA is defined as "the degree to project from 2012 to 2015 on the topic of human-system coto a team or group level. Team SA is defined as "the degree to chain (Lo et al., 2014). In Japan, the Iron and Steel Institute which every team member possesses the SA required for his of Japan (ISIJ) started an academia-industries collaborative or her responsibilities" (Endsley, 1995, p. 39). The success or of Japanfrom (ISIJ) started an academia-industries collaborative to a team or group level. Team SA is defined as "the degree to creative project 2012 to 2015 on the topic of human-system cowhich every team member possesses the SA required for his risk management for establishing resilience against which every team member possesses the SA required for his of Japan (ISIJ) started an academia-industries collaborative or her responsibilities" (Endsley, 1995, p. 39). The success or project from 2012 to on the topic of cofailure of a team onpossesses the success or failure of each of creative project from 2012 to 2015 2015this on project, the topicMizuyama’s of human-system human-system cowhich every teamdepends member the SA required for his risk management for establishing resilience against or her responsibilities" (Endsley, 1995, p. 39). The success or the disturbances. Within group has or her responsibilities" (Endsley, 1995, p.or 39). The success or project from 2012 to 2015 for on the topic of human-system cofailure of a team depends on the success failure of each of creative risk management establishing resilience against its team members. creative risk management for establishing resilience against or her responsibilities" (Endsley, 1995, p. 39). The success or the disturbances. Mizuyama’s failure of aa team OLagainst PMhas AN developed a Within seriousthis game named Cgroup failure ofmembers. team depends depends on on the the success success or or failure failure of of each each of of the creative risk management forproject, establishing resilience its team disturbances. Within this project, Mizuyama’s group has disturbances. Within this project, Mizuyama’s group has failure ofmembers. a team depends on the success or failure oftheeach of the OLPM AN developed a serious game named C its team (C OLLABORATIVE P RODUCTION M ANAGEMENT ), which is a Current industries are forced to act according to rapid its team members. the disturbances. this project, Mizuyama’s group has PM developed aa Within serious game named C OL PMisAN AN developed serious game named ), which COL its team members. (C OLLABORATIVE P RODUCTION M ANAGEMENT Current industries are forced toas act according to and the torapid virtual supply chain of a large-scale make-to-order company changes of circumstances such market changes the developed a Pserious game named ), which COLPMisANaa (C OLLABORATIVE RODUCTION M ANAGEMENT Current industries are forced to act according to the rapid (C OLLABORATIVE P RODUCTION M ANAGEMENT ), which is eta Current industries are forced to act according to the rapid virtual supply chain of a large-scale make-to-order company changes of circumstances such as market changes and to the through makingchain production and delivery decisions (Nonaka change of economicareenvironments. Moreover, the reduction (C OLLABORATIVE PRODUCTION MANAGEMENT ), which is a Current industries forced toas act according to and the torapid virtual supply of aa large-scale make-to-order company changes of circumstances such market changes the virtual supply chain of large-scale make-to-order company changes of circumstances such as market changes and to the through making production and delivery decisions (Nonaka et change of economic environments. Moreover, the reduction al., 2016). Thechain general of this game is twofold; to of expertof workers may bring about drop ofchanges the the work quality, virtual supply of apurpose large-scale make-to-order company changes of circumstances such as market and to the through making production and delivery decisions (Nonaka et change economic environments. Moreover, reduction through making production and delivery decisions (Nonaka et change of economic environments. Moreover, the reduction al., 2016). The general purpose of this game is twofold; to of expert workers may bring about drop of the work quality, games as a platform forand policy development, such as the and theofoccurrences ofbring malfunctions caused by reduction aging of use through making production delivery decisions (Nonaka et change economic environments. Moreover, the al., 2016). The general purpose of this game is twofold; to of expert workers may about drop of the work quality, al., 2016). The general purpose of this game is twofold; to of expert workers may bring about drop of the work quality, use games as a platform for policy development, such as the and the occurrences of malfunctions caused by aging of introduction ofa platform new management forisattaining the equipment may may bring about a drop variety of work unexpected al., 2016). as The general purpose ofconcepts this game twofold; to of expert workers bring about of the quality, use for policy development, such as and the of caused aging of use games games asof a platform for policyconcepts development, such as the the and the occurrences occurrences of malfunctions malfunctions caused byunexpected aging of resilient introduction new management for attaining equipment may bring about a the variety of by and coordinated production scheduling by multiple fluctuations frequently during production. Taking use games as a platform for policy development, such as the and the occurrences of malfunctions caused by aging of introduction new management concepts for attaining the equipment may bring about aa the variety of unexpected introduction of new management concepts forsecondly attaining thea equipment may variety of through unexpected and of coordinated scheduling by multiple fluctuations frequently during production. Taking actors consisting of management the production supply chain, andfor asthe measures suited tobring those about occasions is realized the resilient introduction of new concepts attaining equipment may bring about a the variety of unexpected resilient and coordinated production scheduling by multiple fluctuations frequently during production. Taking resilient and coordinated production scheduling by multiple fluctuations frequently during the production. Taking of hypotheses the production supply on chain, andbehaviors secondly as a measures suited to those during occasions realized through the research tool coordinated to test human within human-system co-creative safety management for actors resilientconsisting and scheduling by multiple fluctuations frequently theis production. Taking actors consisting of the chain, and secondly as measures suited to those occasions is realized through the actors consisting of hypotheses the supply supply chain, andbehaviors secondlywithin as aa measures suited to those occasions is of realized throughThis the research tool to test on human human-system co-creative safety management for the supply chain management system. establishing resilience against a variety disturbances. actors consisting of hypotheses the supply on chain, andbehaviors secondlywithin as a measures suited to those occasions is realized through the research tool to test human human-system co-creative safety management for research tool to management test hypotheses on human behaviors within human-system safety of management for the supplytool chain system. establishing a variety disturbances. This situation is resilience also co-creative true against and becoming serious in the steel research towe testuse hypotheses on human within human-system co-creative safety of management for In the supply chain management system. establishing resilience against aa variety disturbances. This PMAN game to this paper, this testbed of COLbehaviors the supply chain management system. establishing resilience against variety of disturbances. This situation is also true and becoming serious in the steel production industries. Steel production of a the complex the supply chain management system. establishing resilience against a variety consists of disturbances. This In OLPM AN game this paper, we use this testbed of C situation is also true and becoming serious in steel analyze team situation awareness developed during a series situation isindustries. also trueSteel and production becoming consists serious of in a the steel In this paper, we use this testbed of COLPMAN game to production complex to AN game to this paper, we use this testbeddeveloped of COLPM situation isindustries. also trueSteel and production becoming consists serious of in a the steel In analyze team situation during a series production complex production industries. Steel production consists of a complex In AN game to this paper, we use awareness this testbeddeveloped of COLPM analyze team situation awareness during a series analyze team situation awareness developed during a series production industries. Steel production consists of a complex Copyright © 2016 IFAC 548 analyze team situation awareness developed during a series Copyright © 2016, 2016 IFAC 548 Hosting by Elsevier Ltd. All rights reserved. 2405-8963 © IFAC (International Federation of Automatic Control) Copyright © 2016 IFAC 548 Copyright 2016 responsibility IFAC 548Control. Peer review©under of International Federation of Automatic Copyright © 2016 IFAC 548 10.1016/j.ifacol.2016.10.617
2016 IFAC/IFIP/IFORS/IEA HMS 538 Aug. 30 - Sept. 2, 2016. Kyoto, Japan
Tetsuo Sawaragi et al. / IFAC-PapersOnLine 49-19 (2016) 537–542
of game sessions performed by a couple of teams, each of which consists of multiple players. Based on the analysis of players’ decisions and interaction behaviours, some key issues for team situation awareness are identified, and then a novel constructive approach to replaying behaviours of the players are introduced. That is, a simulation model for generating such multiple players’ behaviours within a variety of teams is proposed. This is based upon a descriptive decision model of Garbage Can Model (Cohen et al., 1972), and findings discovered during the empirical analysis of gaming sessions are verified and generalized through a model-based constructive approach.
2. OVERVIEW OF THE GAME COLPMAN is a serious game of supply chain management for steel production. Players of the game consists of the five players; a headquarter accepting orders from customers, an upstream factory producing materials, and three downstream factories processing materials into products that are to be delivered to the customers. This game operates a virtual supply chain of a large-scale make-to-order company through making production and delivery decisions. Details of the game are provided in Nonaka et al. (2016) of this session.
When the schedules are fixed, productions in each factory start and materials produced in a upstream factory are delivered to down stream factories, while downstream factories continue their productions using the supplied materials producing the final products delivered to customers. Wherein, a variety of costs are defined; costs for stocks, costs for delivery, costs for set-up change, and penalties for late delivery. Five players should collaborate to make their decisions on making production schedules and ordering by communicating with each other to maximize sales volumes and minimizing costs and penalties. The game platform consists of decision-making sessions and a simulating part that simulates productions in factories and flows of goods resulted by the former decisions. In the simulation part, a variety of fluctuations such as the ones of lead times, occurrences of manufacturing defects and arrivals of express orders are embedded stochastically. Therefore, each player has to carefully adjust and modify their ongoing plans by communicating the other players in necessity. The game session consists of a series of trials, where each trial is composed of four terms and each term is divided into four periods. At the beginning period of each term, regular orders from customers come to headquarter and all five players start scheduling the production by making material orders in discussing each other, and when they are fixed they are sent to the simulating part that simulate production and deliveries period by period. Because of accidental occurrences of fluctuations, each player has to adjust their decisions at the beginning of the subsequent period in monitoring the production status.
3. COGNITIVE TASK ANALYSIS OF THE GAME
Fig.1 The flow chart of the COLPMAN game Figure 1 illustrates flows of goods and information among the players. Game starts when headquarter accepts orders from customers. An order is defined as a triple consisting of customer, type (of five types), and size (of five sizes) of goods to order. Customers are assumed to be located in places with different distances from downstream factories which supply products to the customers, i.e., times needed to deliver products to customers are different due to a distance gap. Then, headquarter makes a rough plan by assigning downstream factories to produce ordered goods as well as determining appointed dates of delivery of each order. Each of assigned downstream factories makes its own schedule of producing the assigned products as well as making orders of the materials needed for that production to an upstream factory. Orders from the downstream factories to the upstream factory are specified as a tuple of quantity and types of material. Then, accepting orders from downstream factories upstream factory makes its schedule to produce materials to be supplied to the downstream factories. Both of upstream and downstream factories have some amounts of stocks of materials and products, respectively, and should manage these not to produce excess stocks. 549
Despite the large volume of researches on situation awareness across domains, there proposed numerous definitions of SA. One of the most accepted definitions of situation awareness is that of Endsley (Endsley, 1995), called three-level model. Endsley extended a single person’s SA into team SA, where teams are recognized as unit of analysis and team SA is more than a simple sum of individual SA. Related to this, Endsley also defined shared situation awareness as "the degree to which team members possess the same SA on shared SA requirements" (Endsley & Jones, 1997). There are information requirements that are relevant to multiple team members, and a major part of teamwork involves the area where these SA requirements overlap. During COLPMAN game sessions performed by five human players, each player has a subgoal pertinent to his/her specific role that feeds into the overall team goal. Associated with each member's subgoal a set of SA elements about which he/she is concerned exists. As the members of a team are essentially interdependent in meeting the overall team goal, some overlap between each member's subgoal and their SA requirements will be present. To measure and visualize team SA acquired during the game sessions, we introduce a function hierarchy for the game, in which a set of SA elements to be shared with others are structurally configured with different abstraction levels to analyze the shared
2016 IFAC/IFIP/IFORS/IEA HMS Aug. 30 - Sept. 2, 2016. Kyoto, Japan
Tetsuo Sawaragi et al. / IFAC-PapersOnLine 49-19 (2016) 537–542
Fig.2 A functional hierarchy for the the COLPMAN game
539
and incurred in the trial, respectively. Individual trials were tried in one day with no time limit, and subsequent trials were done on different days by keeping regular time intervals. Assignments of the roles within the teams were made at random initially and were kept fixed during the trials till the end of the games. The scores of the game for the teams were recorded for each trial, and the game was reset whenever the teams started new trials. Before starting the game sessions each team was allowed to perform the game as many as three trials to understand the gaming procedures, when the collaborative learning among the players was inhibited. In the gaming sessions, participants are allowed to communicate with other teammates as they like, and logging data of the conversation made during the sessions were all recorded using multiple video cameras. The recorded conversations of the two teams were analysed for each trial. Utterances of the individual players were traced each by each, and they were mapped onto the appropriate nodes within their individual function hierarchies. Thus, a series of function hierarchies in which some parts of the hierarchy were marked highlighted representing temporal single SA of individual players were obtained at the ends of every trial. Then, evolutions of single SA for every player as well as maturity of team SA of the teams were analysed for comparison.
elements of SA. This is illustrated in Fig.2. A function hierarchy is organized in three major abstraction layers, where a top-most layer represents goals to be attained for a successful game play and a second layer consists of the subgoals to attain the goals. Bottom layer represents a set of operations to realize the respective subgoals in the second layer. This hierarchy is defining relationships among the SA elements connected by means-ends relations that are to be shared among the players of the team. In this work, we use this function hierarchy as a common template for specifying the individual SA of each player as well as for evaluating shared SA among the players. The conversations, i.e., protocols, made during the game sessions as well as the game logging data showing the players’ decisions and status of the productions are all recorded. From these data, it is identified which parts of the function hierarchy an individual player is aware of in each stages of the game sessions. Therefore, by highlighting the parts of the hierarchy in awareness, each player’s cognition vector can be defined as his/her individual SA. Moreover, different players’ cognition vectors can be used to evaluate similarities and overlaps of their cognitions, thus shared SA can be quantified dynamically as the game sessions proceeds.
4.2 Results of Experiments A. Evolutions of Team SA: Throughness Attained by Teams Figure 3 shows the scores of the game obtained by team A and team B as the game sessions proceeds as many as six trials; connected line plots show the score obtained and vertical bar charts represents the number of items, i.e., nodes of function hierarchies, at the levels of subgoals and subsubgoals that are shared by more than two players within the team at the end of trials. Each bar chart is divided into different colored parts according to the abstraction levels of a hierarchy, where numbers shows items shared by players.
4. EXPERIMENTS OF GAME 4.1 Overview of Experiments At first we made a series of experiments using the game COLPMAN and investigated into how team SA evolves as the team consisting of five human players iterates and accumulates the gaming sessions. For this purpose, ten male postgraduate school students participated in the experiments, who have no expertise on supply chain management of steel productions. These ten participants were divided into two teams, team A and team B.
As for the scores attained by the teams, team B is achieving a better result early in the trials as compared with team A, while team A achieves worse in the beginning and then gradually improves the score and finally reached to the same level of the score with team B. The number of items shared within the teams and their evolutions are quite different between the two teams; team B has much shared SA among the players even from the early stages of the trials, while
Each team was requested to perform the game sessions with the same teammates for six trials, where a single trial consists of four terms. Participants were requested to gain the score as higher as possible, where the score was calculated from sales volumes and cost/penalties attained 550
2016 IFAC/IFIP/IFORS/IEA HMS 540 Aug. 30 - Sept. 2, 2016. Kyoto, Japan
Tetsuo Sawaragi et al. / IFAC-PapersOnLine 49-19 (2016) 537–542
team A needs more trials to get to share SA within the team. The similar results were obtained as for the comparison with respect to the items corresponding to the bottom layer in the hierarchies.
The results show that in team A the lengths of the fragments are increasing as the trials go on, but the numbers of players contributing to that problem solving are less as compared with team B. On the other hand, in team B conversations lasted longer and more players are contributing that problem solving. 4.3 Findings Obtained from Experiments From the results of the experiments mentioned in the previous section, aspects of team SA for the two teams are quite different and are characterized as follows. Team A: The averaged efforts contributed by the players are less and the team is apt to solve problems on the items that are located in lower portions of the function hierarchy preferring solving operational items to sharing more fundamental issues that are located in the upper parts of the hierarchy.
Fig.3 Results of experiments
B. Evolutions of Performance: Efficiency of Problem Solving by Teams
Team B: The averaged efforts contributed by the players are significant with respect to the lengths of the conversation fragments as well as to the number of players who contributed to collaboration. The team prefers sharing items located in the middle of the function hierarchy to the operational ones in the bottom.
More detailed analysis on the protocols obtained during the game sessions was made to see why the resulted team SA was different between the teams. For this purpose, a chronological series of utterances were segmented into fragments in which utterances were made on the continuous related topics. By tracing the chronological order of the utterances by the players and checking the contents of those, if the transitions of the utterances were on the connected parts of the items within the hierarchies, those were regarded as parts contributing to problem solving on the common issues, while the transitions among the apart portions of the hierarchies are regarded as turnovers of the problems that the team attempted to solve. By tracing the conversations starting from the triggering utterances and segmenting these into a set of fragments according to the above rules, the performances of the two teams were characterized.
5. CONSTRUCTIVE APPROACH TO TEAM SA USING GARBAGE CAN MODEL 5.1 Cohen’s Garbage Can Model The findings mentioned in the previous section were results specific to the two teams who participated in the experiments. Then, we attempt to generalize the empirical findings obtained so far so that we could gain more generalized insights as for the team SA and for designing good teams with respect to SA. For this purpose, we take a constructive approach. The constructive approach is a scientific methodology in which an objective system is to be understood by constructing the system and operating it. We can identify two types of methodologies in simulation studies, one is realistic simulation and the other is constructive simulation. The former tries to make operational copies of actual phenomena as possible as realistic in order to predict what occur in the target phenomena. The latter constructs rather simplified non-realistic models in order to extract essential features and to understand underlying mechanisms and logics of the objective phenomena. In this work, we present a simulation testbed by which human dynamic interactions within the team members during COLPMAN game sessions based on the second simulation concept.
Fig.4 Evolution of utterances during each trial Figure 4 illustrates the results of the number of utterances that were exchanged among the players for solving common problems collaboratively, i.e., lengths of the fragments contributing to the same issues. This graph also illustrates a number of players who joined to collaborative problem solving, i.e., a number of players who participated in the common problem solving in their utterances.
For this purpose, we introduce Cohen’s Garbage Can Model (GCM) that was originally introduced in a field of organization theory for explaining anarchy of organizational behaviors. We extend the original model for finding out the bottlenecks in which the Efficiency and Throughness TradeOff (ETTO) principle of the team is preserved and/or broken by varying a set of essential parameters determining the team characteristics (Hollnagel, 2009). 551
2016 IFAC/IFIP/IFORS/IEA HMS Aug. 30 - Sept. 2, 2016. Kyoto, Japan
Tetsuo Sawaragi et al. / IFAC-PapersOnLine 49-19 (2016) 537–542
The principal components of the GCM are as follows: problems that are concerns of people inside the team, choice opportunities that are occasions when a team is expected to produce behavior that can be called a decision, and participants as human resources of the team contributing to decision-making.
541
evolve according to which access strategy is adopted. ppotential = 1.0 means that a team always takes an access strategy of decision structure A, while ppotential = 0.0 means that a team always takes an access strategy of decision structure B. A decision structure A and B are set corresponding to the findings observed during the game session experiments on team A and B, respectively, mentioned in section 4.
Participants as well as problems are thrown into choice opportunities that are modeled as “garbage cans”, and when the total amount of energy thrown into the garbage can exceeds the energy that is required to solve the problems thrown into that garbage can, that choice opportunity is regarded as resolved, and then they disappear from the organization. The constraints on which problem and which participant is thrown into which choice opportunities can be varied (i.e., an access structure and a decision structure, respectively). Under these constraints, problems as well as participants are allowed to be allocated into one of the choice opportunities, but even after they are thrown into to a particular choice opportunity, they are allowed to move to another choice opportunity anytime whenever some criteria is satisfied. That is, problems may disappear from that choice opportunity and participants may come and go. In this way, the garbage can model can simulate a descriptive organizational behavior of problem solving.
5.3 Results of Team SA Simulations Using GCM
5.2 Mapping COLPMAN game into GCM In our modeling, problems of GCM correspond to the items making up the functional hierarchy of COLPMAN game, and we regard individual team players as participants of GCM. Choice opportunities of GCM correspond to a set of fragments segmented from a series of conversations made during the game sessions. An access structure, i.e., a constraint between choice opportunities and assignable problems, is defined according to the correspondences between fragments of the conversation and pre-classified groups of items in the functional hierarchy that are referred to during the conversation.
Fig.5 Results of GCM simulations Figure 5(a) shows how distributions of different types of problem solving, i.e., decision by resolution, decision by oversight, and decision by flight, vary according to a parameter of ppotential. A type of decision by resolution shows a choice that is made at an opportunity after all contained problems are resolved, and a type of decision by flight shows a choice that is made at an opportunity in spite of some problems not yet made resolved. A type of decision by oversight shows a choice opportunity is set up by some players but no one commit to that then disappears from the subsequent conversation for a while. Figure 5(b) shows how much efforts of the players are needed to reach to problem solving by resolution on average, where efforts are indexed as a number of steps of players’ commitments. Other results showing how the team SA evolves concerning with a number of players who shared the cognition are obtained as well (i.e., derived from players’ cognitive vectors). All the simulation results are averaged results gained from simulations iterated as many as 1000 times.
In addition to a set of original parameters we add a novel parameter of each player’s cognition vector showing his/her temporal awareness on the items of the hierarchy, which is stochastically updated when game players commit to a choice opportunity and the items of that choice opportunity are regarded as cognized by that player. As for a constraint of decision structure, i.e., a constraint between choice opportunities and their committable players, we define two variations as follows. Decision Structure A: This constraint represents that players are apt to selectively commit to the choice opportunity where problem solving by resolution is mostly expected. Decision Structure B: This constraint represents that players are apt to selectively commit to the choice opportunity where the more players are already committed.
5.4 Towards Designing Good Teams
We define a probabilistic parameter ppotential meaning a probability that a team selectively adopts their access strategies according to the decision structure A or B, and by varying this parameter we simulate how the team behaviors
In addition to replaying the observed behaviors of our empirical experiments, we investigate into how our simulation model is used for the purpose of team design. There would be so many factors that can affect the team 552
2016 IFAC/IFIP/IFORS/IEA HMS 542 Aug. 30 - Sept. 2, 2016. Kyoto, Japan
Tetsuo Sawaragi et al. / IFAC-PapersOnLine 49-19 (2016) 537–542
performances, some of which are identified as a number of parameters defined in the Garbage Can Model, such as selection of two constraints, selection of participants, distribution of the expertise among the participants, etc. Team performance should be evaluated with respect to their efficiencies, thoroughness, depth and broadness of team SA, etc.
division of roles within the team, homogeneity and/or heterogeneity of tam members with respect to their expertise, etc., but our proposal of constructive approach does contribute to discovering a novel way for team fostering and for human resource management.
6. CONCLUSIONS In this work, analysis of team situation awareness using serious game of supply chain management for steel production was provided. Especially, we focus into the individual players’ mental models and we introduced a function hierarchy tree to visualize players’ temporal mental models through cognitive analysis. Through a comparative analysis of the two different teams with respect to their shared team situation awareness, we discussed about the relationships between the maturity of the shared mental models and performance scores attained by the teams. Then, a simulation model for generating such multiple players’ behaviours within a variety of teams was proposed. We showed findings discovered during the gaming sessions are verified and generalized through a model-based constructive approach, and discussed about its extension towards developing a novel approach to team design.
Fig.6 Simulated results when external burdens are increased for a virtual team having decision structure A In spite of many freedoms for team design, the essence is that a team behaviour is not the single sum of individual decision-makers or problem solvers, but is rather an “open” and “ambiguous” process, and the management for such a complex system will be a critical issue especially when the team faces with unexpected disturbances. Human dynamic interactions within the team are determined not only specifications of the team per se but also by what are externally provided to the team and how the team adapts to such external disturbances.
REFERENCES Cohen, M.D., March, J.G. and Olsen, J.P. (1972). A Garbage Can Model of Organizational Choice, Administrative Science Quarterly, 17(1). Endsley, M.R. (1995). Toward a Theory of Situation Awareness in Dynamic Systems, Human Factors, 37(1), pp.32–64. Endsley, M. & Jones, W. (1997). Situation awareness, information dominance, and information warfare, Wright-Patterson AFB, OH: United States Air Force Armstrong Laboratory. Hackman,J.R. (1987). The design of work teams, Handbook of organizational behavior, pp.315-342. Hollnagel, E. (2009). The ETTO principle: efficiencythoroughness trade-off: why things that go right sometimes go wrong. Aldershot, Hants, England: Ashgate. LoJ.C., van den Hoogen, J. Meijer, S.A. (2014). Testing Changes in Railway System Through Gaming Simulation: How Different Types of Innovations Affect Operators’ Mental Models Proceedings of the 5th International Conference on Applied Human Factors and Economics AHFE 2014. Nonaka, T, Miki, K., Odajima, J. and Mizuyama, H. (2016). Analysis of Dynamic Decision Making Underpinning Supply Chain Resilience: A Serious Game Approach, to be presented at IFAC-HMS 2016. Salas, E., Prince, C., Baker, D.P., Shrestha, L. (1995). Situation Awareness in Team Performance: Implications for Measurement and Training, Human Factors, 37 (1), pp.123-136.
In this work, using the developed simulation system for COLPMAN game we investigated into the relation between team behaviours and external workloads that are thrown into the game. If we were to discover relations between those, we could find out an indirect way for designing the team by controlling the external workloads. This is a so-called indirect control in contrast to the conventional direct control. The indirect control method guides the system by controlling the system's boundary conditions and is a concept in opposition to the direct control method that controls the parameters in the system. By applying this method to the virtual team, we expect to realize an effective method of guiding the team to evolve into a mature team. Figure 6 shows the results of the simulation when external burdens are varied for a team characterized with the feature of decision structure A. The results show that problem solving efficiencies are degraded due to the difficulties of problem solving thrown into the team. However, more minute investigation into how their team SA evolves, the result shows both the depth and the broadness of team SA are improved more as compared with team SA cultivated for less burdens. This shows a possible way of nurturing teams by controlling the provision of external workloads. The result may indeed depend upon the team characteristics,
553