- Email: [email protected]
A knowledge-based decision support system for shipboard damage control
Expert Systems with Applications 39 (2012) 8204–8211
Contents lists available at SciVerse ScienceDirect
Expert Systems with Applications journal homepage: www.elsevier.com/locate/eswa
A knowledge-based decision support system for shipboard damage control F. Calabrese c, A. Corallo b, A. Margherita b,⇑, A.A. Zizzari a a
Centro Cultura Innovativa d’Impresa, University of Salento, Campus Ecotekne, Via Monteroni s.n., 73100 Lecce, Italy Department of Innovation Engineering, University of Salento, Campus Ecotekne, Via Monteroni s.n., 73100 Lecce, Italy c Apphia s.r.l., via Clementina Carrelli n. 26, 73100 Lecce, Italy b
a r t i c l e
i n f o
Keywords: Damage control system Decision support system Expert system Kill card Knowledge-based system Shipboard management
a b s t r a c t The operational complexity of modern ships requires the use of advanced applications, called damage control systems (DCSs), able to assist crew members in the effective handling of dangerous events and accidents. In this article we describe the development of a knowledge-based decision support system (KDSS) integrated within a DCS designed for a national navy. The KDSS uses a hybrid design and runtime knowledge model to assist damage control operators through a kill card function which supports damage identification, action scheduling and system reconfiguration. We report a fire fighting scenario as illustrative application and discuss a preliminary evaluation of benefits allowed by the system in terms of critical performance measures. Our work can support further research aimed to apply expert systems to improve shipboard security and suggest similar applications in other contexts where situational awareness and damage management are crucial. Ó 2012 Elsevier Ltd. All rights reserved.
1. Introduction Navy ships have been traditionally manned with a large crew involved in the manual control of onboard systems. The reduction of through-life costs of vessels is today a priority and there is an interest towards reducing crew without affecting damage control capabilities or jeopardizing the ability of ships to complete their missions (Cosby & Lamontagne, 2006). Beside efficiency pressures, factors like the increased complexity of modern vessels, the requirements for easy maintainability and more sophisticated operational demands generate a need for intelligent functionalities and leading-edge technologies (Bøgh & Severinsen, 2009) which can assist human decisions and actions onboard. In this endeavor, one relevant area is related to the management of events which may lead to shipboard damage and crew danger. These events require rapid actions, also without on-site human intervention, to prevent serious injuries to personnel or damages to vital ship systems. Whereas damage control has been traditionally a manual and manpower-intensive function, the automation of emergency management operations is today driven by complex technology architectures called damage control systems (DCSs) and related progresses in human-system integration, which gets increasing attention in ship design (Runnerstrom, 2003). A DCS is an information-retrieval and equipment-control system that gives ship personnel the ability to detect, analyze, and handle ⇑ Corresponding author. Address: University of Salento, Department of Innovation Engineering, c/o Euro-Mediterranean Incubator, Campus Ecotekne, Via Monteroni s.n., 73100 Lecce, Italy. Tel.: +39 0832 297922; fax: +39 0832 297927. E-mail address: [email protected] (A. Margherita). 0957-4174/$ - see front matter Ó 2012 Elsevier Ltd. All rights reserved. doi:10.1016/j.eswa.2012.01.146
various types of damage situations, based on the collection and processing of vast quantities of shipboard information. In the navy context, a damage control system (DCS) is aimed to assure the timely and informed application of men and equipment in scenarios such as fire or flooding, violation of the ship closure state, threats to essential equipments, ventilation close down, and atomic/biological/chemical issues. DCSs are also relevant for emergency training and damage instructor assistance purposes (Bulitko & Wilkins, 1999; Peters, Bratt, Clark, Pon-Barry, & Schul, 2004). The noteworthiness of these systems is proven by the number of leading market players (e.g. ABB, L3, Northrop Grumman, Rockwell, and Siemens) involved in the design and development of innovative solutions for damage control. The assistance to damage control operators, with recommendations for counteractions and reconfigurations, requires a highly structured approach to problem identification and action planning. The field of expert and decision support systems can thus provide a relevant contribution to design more performing DCSs. However, the study of expert systems and DSS in navy contexts has mostly focused on the design process whereas a very limited number of contributions have addressed the implementation of integrated systems to ensure the safety and operational stability of modern ships. In this paper, we show the development of a knowledge-based decision support system (KDSS) which has been integrated within the DCS designed for the operating needs of a national navy. We start from the analysis of the typical damage control process and identify a model of knowledge acquisition and reuse in damage management scenarios. The model is implemented through the development of a kill card function providing an interactive
F. Calabrese et al. / Expert Systems with Applications 39 (2012) 8204–8211
interface and a shared decision and action platform for damage control personnel onboard. The tool provides a graphical information-retrieval and equipment-control dashboard that gives damage crew the ability to handle various types of damage control situations. The remainder of the paper is structured as follows: Section 2 reviews existing literature on damage control and DSS applications in the navy context; Section 3 introduces the research requirements and describes the overall architectural design; Section 4 illustrates the KDSS developed; Section 5 presents an illustrative application related to a fire-fighting scenario and proposes a preliminary evaluation of quali-quantitative benefits; Section 6 concludes the paper and draws avenues for future research. 2. Damage control and DSS The concept of damage control is not limited to the navy industry. In fact, it is largely used in the field of medicine, surgery, and injury prevention as well as in project management, politics, media, and industrial production. In general, the term is used to describe the process of identifying, monitoring and dealing with any problem that may jeopardize a given system or endeavor. A first area in which information and decision support systems have been developed for damage control purposes is the field of dynamic emergency response and resource intervention (Minciardi, Sacile, & Trasforini, 2007; Turoff, Murray, de Walle, & Yao, 2003). The use of expert systems and virtual reality to support decision making in emergency management situations has been also studied (Beroggi, Waisel, & Wallace, 1995; Caro, 1992). Other contributions have investigated the use of computer-based systems to support human improvisation in extreme events (Mendonca, 2007) and the inter/intra-organizational communication and coordination in emergency response (Kanno & Furuta, 2006). As a specific area of application, the response logistics to roadway network incident (Zografos, Androutsopoulos, & Vasilakis, 2002) has been studied. Besides emergencies, damage management has represented an area of interest for expert systems and neural networks in manufacturing and engineering contexts, with a focus on structural damage detection and assessment (Barai & Pandey, 2000; Jiang, Zhang, & Zhang, 2011; Mehrjoo, Khaji, Moharrami, & Bahreininejad, 2008; Ubeyli & Ubeyli, 2009). Despite the relevance of the topic and the applicability of decision support and expert system principles, few contributions have instead focused on the development of intelligent systems for shipboard damage control. An effective damage control system (DCS) needs a systemic approach to realize key functions such as provide support to control personnel to make informed and real-time decisions, enhance total ship, coordinated, real-time control of men and equipment at the scene of damage, and allow changeover from remote-automated to local-manual control in case of emergency (Geer, 1988). In the area of intelligent applications for shipboard damage control, an expert system was created to support the cognitive processes involved in ship piloting and collision avoidance (Grabowski & Wallace, 1993). With a more specific focus on damage management, a rule-based expert system based on information from navy damage control tactics, procedures, doctrine, and experts was presented (Tate, 1996). A fuzzy distributed expert system was built to assist command and control activities (Simoes-Marques & Pires, 2003) and a virtual environment has been developed to support emergency planning decisions by considering what could occur when fluids disseminate through ship compartments, such as flooding, fire, or contamination (Varela & Guedes Soares, 2007). Finally, an expert system was developed for ship auxiliary machinery troubleshooting (Cebi, Celik, Kahraman, & Deha Er, 2009).
8205
The design and development of expert and decision support systems in ship contexts has been mostly focused on the design process (Park & Storch, 2002). Different contributions have addressed the aided design of ship systems automation (Arendt, 2004; Arendt & van Uden, 2011; Kowalski, Arendt, Meler-Kapcia, & Zielnski, 2001; Kowalski, Meler-Kapcia, Zielinski, & Drewka, 2005), the support of the conceptual design stage based on knowledge engineering (Lee, 1999; Lee & Lee, 1999), and the analysis of design problems and assessment of trade-offs between performance and cost (Chou & Benjamin, 1992). Two specific studies have developed an expert system to support compartment design of a crude oil tanker (Lee & Lee, 1997) and a DSS for vessel fleet scheduling (Fagerholt, 2004). All these studies can provide a larger background to define the requirements for ship design, building and management, with the purpose to extend the research on decision support systems applied for ship personnel security and control of casualties to ship systems. 3. Overview of the DCS The development of our KDSS was framed within a collaborative project with Avio SpA. This a leading company operating both in the aerospace propulsion (participates in military programs such as the F-35 JSF and civil partnerships with General Electric, Pratt & Whitney, and Rolls Royce) and in the marine industry for the provision of control, automation and propulsion systems and components (e.g. turbine control, lubrication and fuel systems). The company is a Marine System Supplier (MSS) of General Electric. In the last years, Avio has worked in the development of innovative damage control systems (DCSs) for national navies. The DCS is a supervisor system embedded in the Integrated Platform Management System (IPMS), which is a distributed hardware architecture used for real-time monitoring of the ship propulsion, mechanical, electrical, auxiliary and damage control systems. Monitored components include gearboxes, pitch propellers, power generation sets, power distribution switchboards, electrical distribution centers, fire pumps, systems for heating, ventilation, air conditioning, chilled water, and so forth. In practice, the IPMS controls all the onboard equipment, excluding weapons/sensors (for military ships) and the ship’s communication and navigation equipment. The general IPMS architecture comprises Multi-Function Consoles (MFCs) and Remote Terminal Units (RTUs). MFCs are mostly laptops and workstations providing the human–machine interface for the operators at various shipboard locations whereas RTUs are used for data acquisition and control and they are connected to sensors and actuators (e.g. FDS – fire detection sensors, pumps, fans). The IPMS is endowed with a runtime application allowing to monitor the whole ship from each MFC. Whereas the IPMS represents the hardware backbone for damage control operations, the DCS is the software platform configured within the IPMS with functions such as monitoring of ship subsystems, longitudinal, planar and isometric views, Tiled Layered Graphics (TLGs) approach (for automatic de-cluttering and display of complex information), casualty management, support to manage emergency states, event log and report, and compartment monitoring. These functionalities are integrated into four modules: (1) Damage Control Management System (DCMS), which enables to automatically acquire all the relevant ship safety and other data needed to handle damages, display data to the operator in an optimized way, handle alarms, and rapidly share/communicate the information between the different MFCs; (2) On Board Stability Software (OBSS), to obtain and visualize ship stability data (e.g. tanks
8206
F. Calabrese et al. / Expert Systems with Applications 39 (2012) 8204–8211
Fig. 1. IPMS hardware architecture and main DCS interface.
level, flooding sensors, water tight doors status, etc.) and calculate the stability parameters; (3) Closed Circuit TV (CCTV), which activates cameras to monitor real-time the compartments of the ship; and (4) Decision Support System (DSS), to assist the damage control officer in case of critical events by indicating the most suitable procedures to handle the specific situation. The four modules are organized within an interactive interface which allows the operator to access each function starting from a main window and navigating within the system by means of guided paths and hyperlinks. Fig. 1 shows the architecture of the IPMS and the basic interface of the DCS. The red arrow indicates that the DCS interface can be accessed from every damage control terminal (the RTU) onboard. To develop the KDSS, we first studied the decision/action process of the damage control operator onboard. We have then identified the information flows at the basis of decisions ad actions and developed the four core components of the DSS (Bonczek, Holsapple, & Whinston, 1981), i.e. the knowledge system (sources of data and information), the language system (input format), the presentation system (interface and layout) and the problemprocessing system (software engine). Next section describes the output of our work.
4. KDSS for shipboard damage control Damage and emergency management efforts onboard are coordinated by a damage control officer (DCO), which is usually supported by a damage control assistant (DCA) and a team in charge of maintaining situational awareness and taking actions to prevent injury to personnel, damage to ship systems, or loss of the ship as well. The KDSS developed has a direct impact on the decision making process and actions of these actors. A generic decision making flow includes the recognition of the problem, the listing of objectives, the perception of environment and constraints, the listing of options, the decision analysis and the action plan (Arbel & Tong, 1982). In the damage control perspective, these steps are specifically translated in a flow including five key steps (Fig. 2). The monitoring of ship systems through the damage control platform allows the operator to acquire situational awareness of damage (step one) and identify where the damage is located and which is its extent (step two). Next, the operator can start a set of preliminary actions aimed to contain and control the effects of damage (step three) and activate the damage control systems and crew to eliminate the causes of emergency and prevent further
Fig. 2. Decision and action flow of damage control operator.
F. Calabrese et al. / Expert Systems with Applications 39 (2012) 8204–8211
issues (step four). Finally, the operator reconfigures the navy and control systems after that the problem has been completely solved (step five). Throughout the five steps, the operator takes critical decisions related with what to do, what to do first, how to undertake actions, and in which order undertake actions. In order to streamline the decision flow and the execution of actions, we identified four major requirements to drive the design of the KDSS: (1) monitor all damage control status and operations at any time and from each control position onboard; (2) support damage control operations by acquiring all ship’s and security relevant data; (3) allow efficient presentations of information to the damage control operator; (4) provide decision aids, actions, procedural checklists and alarms when handling emergency situations. A KDSS is a special system for decision support which is able to recommend actions based on specialized problem-solving expertise stored as facts, rules, procedures, or in similar structures. We first designed the knowledge model underlying the system. Basic knowledge to support the decisions and actions of the damage control crew derives from two main sources: (1) design time knowledge sources, which include the ship and damage control data available at the design of the system; and (2) run time knowledge, including damage control data received ‘‘in process’’ through the damage control equipments. Design time knowledge was obtained from the ship structure and engineering data (e.g. ship layout and dimensions of compartments), navy/ship rules (e.g. operating and security management procedures) and damage control officer (e.g. engineering expertise, design suggestions, insights). Run time knowledge is implemented in the system through damage system information (e.g. fire and flooding sensors connected with the RTUs) and shared communications among the damage control operators onboard (e.g. separated actions which have to be consolidated into a unique damage checklist).
8207
Concerning the problem processing level of the KDSS, the knowledge sources are integrated within the kill card, i.e. an information and operation support dashboard which groups logically all the information related to the ship equipment and security systems, provides predefined automatic control sequences to respond to specific casualty conditions, and allows the operator to rapidly (knowing and) executing the correct actions at the moment. Since a damage generally originates in one specific compartment of the ship (and then it can propagate to other compartments), each compartment is associated to a dedicated kill card. We have thus developed a total of about 300 cards, which is the average number of compartments for a large vessel. We implemented the KDSS as a Microsoft Windows Environment developed in Visual Studio 2008. We used C# as programming language as it presents the power of C++ and the slickness of Visual Basic. Besides, the language supports the introduction of XML comments which can turn into documentation. The comments are placed into XML format and can then be used as documentation which can include example code, parameters, and references to other topics. Fig. 3 shows the overall model of our KDSS. We developed the kill card function with two main components, i.e. the editor and the viewer. The editor allows to create a new card (with a preview function), modify or delete an existing card from the kill card database if/when basic requirements related to ship structure, rules and operating procedures should change (i.e. the design time knowledge sources). The viewer allows to open and visualize an existing kill card generated through the editor. In order to retrieve a card, the operator has three options: (1) use a plan or isometric view of the ship to click on the description of a compartment; (2) choose a card from a purposeful kill card area by using a database tree structure (which describes the whole ship); (3) use a search function based on compartment names. In the viewer function, the damage control operator can also share the action list, i.e. score when response actions (e.g. in case of fire) have been started and/or completed and transmit this information to every MFC onboard. Finally, it is possible to reset changes by using a clear function and control the dynamic buttons of devices
Fig. 3. Model of the knowledge-based DSS for shipboard damage control.
8208
F. Calabrese et al. / Expert Systems with Applications 39 (2012) 8204–8211
Fig. 4. Kill card structure.
through a device window. All the information contained in every kill card and all the changes made on them are stored in a database using a XML file whose name corresponds to the ship compartment name (e.g. wardroom, engine room). The general format of a kill card (the ‘‘template’’) is an empty format structured in text fields and buttons (allowing to operate various devices) and it includes three sections or areas: (1) compartment identification area, showing critical information such as ship general data and compartment location; (2) summary area, with information related to ship and compartment sensors; (3) detail area, including several tables with information related to ventilation control, danger hazards, fire fighting installations, actions, compartment views and video clips, and status of devices. The damage control operator can visualize actions from a predefined list and ‘‘check off’’ (see Fig. 4) when these actions have been successfully completed. The system also allows to link kill cards of different compartments when the effects of damage propagate into different ship rooms. In this way, it is possible to use the sensors and devices available in more compartments, identify priority areas, execute shared actions, and so forth. The basic structure and content of a kill card template can be modified through a purposeful configurator. The editor, viewer and configurator components are operationalized by a software application available in every MFC of the ship, allowing damage
control operators to be always in possess of the information pertinent to the emergency situation managed at the moment. The kill card ‘‘uses’’ the design information (such as compartments layout and dimensions, ship security management procedures, and damage control officer insights) and run time knowledge (like system status, status of actuators and shared information on actions taken by different operators), to provide expert assistance to the damage control crew. The kill card is the main element of the user interface and it allows a dynamic situational awareness of damage, damage localization and extent, preliminary counteractions, damage control systems and crew activation and reconfiguration. The KDSS suggests therefore to the damage control operator what to do, what to do first, how to undertake actions, and in which order undertake actions in the different emergency situations which may happen onboard. Next section shows an illustrative scenario of fire fighting in which the system can provide its support to damage control operations. A preliminary evaluation of benefits achieved is also described.
5. Illustrative scenario and evaluation of benefits We describe hereafter an illustrative scenario in which a fire emergency onboard requires rapid actions to avoid serious injuries to personnel or damages to vital ship systems. In this case, the damage control crew is involved in the following sequence of steps:
F. Calabrese et al. / Expert Systems with Applications 39 (2012) 8204–8211
(1) acquire awareness about a fire happened onboard; (2) identify where fire is located and which is its extent; (3) start a set of preliminary actions aimed to contain and control the effects of fire; (4) activate damage control systems and crew to eliminate causes of fire and prevent further damages; (5) reconfigure and restore the ship and damage control systems after fire. The five steps are supported by the integrated functionalities allowed by the IPMS and the DCS of the ship. In particular, the KDSS developed has a direct impact on steps 3 and 4. In the first step, the sensors installed in the compartment concerned with the fire transmit an alarm or warning message which is visible in the summary area of the kill card interface. The operator can thus immediately become aware of the emergency. In step two, the system allows the operator to visualize which compartment is concerned and which is the seriousness of damage. These analyses are supported by different visualization options of the compartment and the use of a zoom feature and red signs on the interface (in Fig. 5 a red box in the ship layout indicates the compartment where fire has started). Besides, devices and sensors acquire quantitative measures (e.g. temperature, pressure, CO2 level, etc.) which are sent to the system for real-time damage evaluation. After that, the operator can start a set of preliminary actions (step 3) aimed to contain and control the effects of fire. At this purpose, the operator access the kill card database to retrieve the specific kill card of the compartment concerned with the fire. In fact, this kill card will indicate the list of actions suggested by the KDSS (on the basis of the knowledge model underlying the system) to resolve the problem. An example of action list in case of fire is the following: (1) close ventilation; (2) preserve watertight integrity; (3) maintain vital systems; (4) isolate, prevent, extinguish, combat and remove the effects of fire; (5) facilitate the care of personnel casualties; and (6) make rapid repairs to the ship’s structure and equipment. The operator can then activate the damage control systems and crew (step 4) to prevent further damage and limit current issues.
8209
For example, the operator can monitor the status of devices and control remotely pumps and fans which are present at the fire scene. Then, he/she can assemble the damage control task force and execute the secondary action plan with the purpose to completely extinguish the fire. When the alarm status is over, the operator can then reconfigure and restore the ship and damage control systems. Through the remote control functionalities of the KDSS, the sensors and devices, as well as all the action lists turned on can be reset and made ready in case of a new emergency. The fire fighting case, along with other damage management scenarios, was created to test the behavior of the crew and the system through the damage management process. To validate the KDSS, we used indeed a combination of technical, empirical and subjective methods (Adelman, 1992; Borenstein, 1998; Papamichail & French, 2005). We realized a technical validation by cross-checking the information retrieval and visualization output obtained from different damage control stations onboard, with a specific attention to the consistency of data obtained and knowledge sources used. In particular, we ascertained that all the MFCs on board show the same data to the different damage control operators. We obtained empirical validation through simulations aimed to verify the correct functioning of the tool and measure its performance as enhancer of the decision making and action process. A combination of outcome and process criteria is in fact highly relevant for DSS evaluation (Phillips-Wren, Hahn, & Forgionne, 2004). Empirical validation was strictly linked with the subjective assessment obtained with user confirmation about the benefits/utility and usability of the tool. At this purpose, we interviewed the ship damage officer and operators to understand if the system meets the needs of users and how well the interface was designed. We have also measured the benefits allowed by the DSS on the basis of four critical metrics and four qualitative indicators. Metrics mostly refer to the impact in terms of user and damage control process performance and include: (1) awareness time, i.e. the average time needed by the damage control operator to acquire situation awareness after that a damage event happens; (2) action time,
Fig. 5. Kill card windows in the fire fighting scenario.
8210
F. Calabrese et al. / Expert Systems with Applications 39 (2012) 8204–8211
Table 1 Indicators and benefits allowed by the KDSS. Indicator
Without KDSS
With KDSS
Awareness time Action time Crew need Action cost Navigability Multimediality Information sharing Interoperability
15 min 45 min 5 Operators 112,5$ Not allowed Not allowed Docs, phone, meetings
1 min 15 min 2 Operators 15$ Hypertexts and function trees Audio/video MFCs
Not allowed
Integrated DCS
i.e. the average time needed to activate the key procedures to solve the problem or reduce the damage; (3) crew need, i.e. the average number of total crew members required to act throughout the whole damage handling process; (4) action cost, i.e. the average cost for the complete end-to-end decision and action procedure. Qualitative indicators refer to system features in terms of: (1) navigability, i.e. the direct and fast access to single functions of the tool; (2) multimediality, i.e. the possibility to integrate different forms of information visualization and human–system interaction tools; (3) information sharing, i.e. the availability of the same damage information and actions status at the different control positions onboard; and (4) interoperability, i.e. the integration of different software applications with effects which are immediately implemented within the different tools. Table 1 reports a comparison of indicators in the ‘‘without KDSS’’ and ‘‘with KDSS’’ scenarios. ‘‘Without KDSS’’ values are average measures reported by navy personnel involved in damage control operations whereas after-implementation values (‘‘with KDSS’’) derive from a preliminary estimation based on simulations and managerial/operator validation at the utilization site. The situational awareness time is about 15 min without the system. In fact, a visible damage or problem in a ship system or compartment activates the internal crew communication process through which the warning message is received by the damage control operator. The use of kill cards connected with sensors and cameras allows to reduce such time to few seconds (and less than 1 min) needed by the operator to visualize the automatic warning message on the screen (with an improvement of about 93%). The damage action time can be decreased from 45 to about 15 min (improvement of 67%), as the operator can now immediately start a set of corrective actions directly from the remote terminal unit and there is no need to be physically present at the damage site. The KDSS provides the operator with a kill card which synthesizes all the information and knowledge critical to successfully handle the problem. Naturally, damage action time represents a very critical measure for the safety of personnel and the survival of vital ship systems. Concerning crew need, five persons are needed on average in the damage control staff to handle the situation from end to end. In the with-KDSS scenario, only two persons (60% decrease) are required to use the interface and support the overall damage recognition and recovery action process. As a consequence of reduction in action time and crew need, the total action cost related with the employment of specialized crew (materials, resources consumption and other costs are thus left out in this calculation) can be now reduced from 112,5$ to about 15$, with a reduction of about 87% (we made an assumption of 30$ for the average hour cost of a damage crew member). Moreover, the automation of routine actions allows to further improve the performance of damage control operations since the overall efficiency of operations is increased and damage personnel can be involved in value added activities connected with damage control or other tasks related with shipboard management.
Concerning qualitative benefits, navigability, multimediality, information sharing and interoperability are new features allowed by the KDSS and the integrated damage control platform. Navigability of the tool is supported by the use of hyperlinks and function navigation trees. The system allows some multimedia features with audio and video signals directly coming from ship compartments. Information sharing, which was based on exchange of documents, physical meetings and phone calls, is now supported by the direct communication among the Multi-Function Consoles (MFCs) on board. Finally, the system also supports interoperability as the IPMS, the modules of the DCS and the KDSS are fully integrated and can be customized based on user needs and requirements.
6. Conclusions The identification and management of events that may lead to shipboard damage and crew danger are interesting areas of application for expert and decision support methods and tools. National navies around the world are in fact turning to enhanced and distributed damage control systems to achieve higher level of security and operational efficiency through effective information sharing, fast problem identification and action planning and automation. In this paper, we have presented a knowledge based DSS which uses design time and run time knowledge sources to streamline the decision making process and sequence of actions required to the damage control operator in case of emergency. The system reduces situational awareness time, action time, crew need and overall action cost. The KDSS also allows full navigability of damage control information, the use of multimedia tools for damage monitoring, the interoperability with other DCS applications, and a more effective information sharing. The dedicated displays onboard enable the operators to immediately identify the emergency and initiate corrective actions. The ship-wide data network allows several dispersed damage stations to retrieve coherent information and thereby effectuate a coordinated and effective action, resulting in reduced damage control response time, enhanced consistency of actions, and reduced manning. The use of the application developed could be enlarged to other contexts (e.g. building sites, nuclear and other energy production sites) where the monitoring of risky events in different compartments or operation areas requires advanced control and decision/ action support technologies. In such cases, the KDSS can be of value to increase the situational awareness of damage crew members and enhance data consistency through the use of automatized control devices for the remote identification of risks. Next research will be addressed to extend the application of the DSS for training purposes, and in particular for on-the-job training of damage crew members. The onboard training system of the ship could be indeed used to simulate events in normal ship operation as well as in degraded conditions using the same interface. A second area of development is represented by the adoption of enhanced reality and 3D technologies and functionalities within the system. This could further enhance the situational awareness of operators and their ability to promptly identify and assess the problem, resulting in faster and more effective actions.
Acknowledgments The authors are grateful to the persons at Avio SpA who have collaborated in the design and implementation activities. A particular acknowledgment goes to Marco Rosso for his support in the final revision of the article.
F. Calabrese et al. / Expert Systems with Applications 39 (2012) 8204–8211
References Adelman, L. (1992). Evaluating decision support and expert systems. New York, NY: Wiley. Arbel, A., & Tong, R. (1982). On the generation of alternatives in decision analysis problems. Journal of the Operational Research Society, 33, 377–387. Arendt, R. (2004). The application of an expert system for simulation investigations in the aided design of ship power systems automation. Expert Systems with Applications, 27(3), 493–499. Arendt, R., & van Uden, E. (2011). A decision-making module for aiding ship system automation design: A knowledge-based approach. Expert Systems with Applications, 38(1), 410–416. Barai, S. V., & Pandey, P. C. (2000). Integration of damage assessment paradigms of steel bridges on a blackboard architecture. Expert Systems with Applications, 19(3), 193–207. Beroggi, G. E. G., Waisel, L., & Wallace, W. A. (1995). Employing virtual reality technology to support decision making in emergency management. Safety Science, 20(1), 79–88. Bøgh, S. A., & Severinsen, T. (2009). Damage management control system on the Danish navy ships. In Proceedings of the 14th international ship control systems symposium (SCSS). Ottawa, Canada. Bonczek, R. H., Holsapple, C., & Whinston, A. B. (1981). Foundations of decision support systems. New York, NY: Academic Press. Borenstein, D. (1998). Towards a practical method to validate decision support systems. Decision Support Systems, 23(3), 227–239. Bulitko, V. V. & Wilkins, D. C. (1999). Automated instructor assistant for ship damage control. In Proceedings of the 16th national conference on artificial intelligence. Orlando, FL: American Association for Artificial Intelligence. Caro, D. H. J. (1992). Expert support systems in an emergency management environment – An evaluative framework. Educational and Training Technology International, 29(2), 132–142. Cebi, S., Celik, M., Kahraman, C., & Deha Er, I. (2009). An expert system towards solving ship auxiliary machinery troubleshooting: SHIPAMTSOLVER. Expert Systems with Applications, 36(3), 7219–7227. Chou, Y.-C., & Benjamin, C. O. (1992). AI-based decisions support system for naval ship design. Naval Engineers Journal, 104(3), 156–164. Cosby, L., & Lamontagne, Y. (2006). Critical assessment of damage control system technologies. Defence R&D Canada Report. Fagerholt, K. (2004). A computer-based decision support system for vessel fleet scheduling – Experience and future research. Decision Support Systems, 37(1), 35–47. Geer, D. (1988). Advanced damage control systems. Naval Engineers Journal, 100(3), 143–153. Grabowski, M., & Wallace, W. A. (1993). An expert system for maritime pilots: Its design and assessment using gaming. Management Science, 39(12), 1506–1520. Jiang, S.-F., Zhang, C.-M., & Zhang, S. (2011). Two-stage structural damage detection using fuzzy neural networks and data fusion techniques. Expert Systems with Applications, 38(1), 511–519. Kanno, T., & Furuta, K. (2006). Modeling and simulation of inter-and intraorganizational communication and coordination in emergency response. International Journal of Emergency Management, 3(2–3), 149–167.
8211
Kowalski, Z., Arendt, R., Meler-Kapcia, M., & Zielnski, S. (2001). An expert system for aided design of ship systems automation. Expert Systems with Applications, 20(3), 261–266. Kowalski, Z., Meler-Kapcia, M., Zielinski, S., & Drewka, M. (2005). CBR methodology application in an expert system for aided design ship’s engine room automation. Expert Systems with Applications, 29(2), 256–263. Lee, D.-K. (1999). Hybrid system approach to optimum design of a ship. Artificial Intelligence for Engineering Design, Analysis and Manufacturing: AIEDAM, 13(1), 1–11. Lee, K.-H., & Lee, K.-Y. (1997). Knowledge-based nonmonotonic reasoning process in ship compartment design system. Expert Systems with Applications, 13(2), 145–154. Lee, D., & Lee, K.-H. (1999). An approach to case-based system for conceptual ship design assistant. Expert Systems with Applications, 16(2), 97–104. Mehrjoo, M., Khaji, N., Moharrami, H., & Bahreininejad, A. (2008). Damage detection of truss bridge joints using artificial neural networks. Expert Systems with Applications, 35(3), 1122–1131. Mendonca, D. (2007). Decision support for improvisation in response to extreme events: Learning from the response to the 2001 world trade center attack. Decision Support Systems, 43(3), 952–967. Minciardi, R., Sacile, R., & Trasforini, R. (2007). A decision support system for resource intervention in real-time emergency management. International Journal of Emergency Management, 4(1), 59–71. Papamichail, K. N., & French, S. (2005). Design and evaluation of an intelligent decision support system for nuclear emergencies. Decision Support Systems, 41(1), 84–111. Park, J.-H., & Storch, R. L. (2002). Overview of ship-design expert systems. Expert Systems, 19(3), 136–141. Peters, S., Bratt, E. O., Clark, B., Pon-Barry, H., Schul, K. (2004). Intelligent systems for training damage control assistants. In Interservice/industry training, simulation, and education conference (I/ITSEC). Phillips-Wren, G. E., Hahn, E. D., & Forgionne, G. A. (2004). A multiple-criteria framework for evaluation of decision support systems. Omega-International Journal of Management Science, 32(4), 323–332. Runnerstrom, E. (2003). Human systems integration and shipboard damage control. Naval Engineers Journal, 115(4), 71–80. Simoes-Marques, M. J., & Pires, F. J. (2003). SINGRAR – A fuzzy distributed expert system to assist command and control activities in naval environment. European Journal of Operational Research, 145(2), 343–362. Tate, D. L. (1996). Development of a tactical decision aid for shipboard damage control. NRL, Report NRL/FR/5580-96-9837. Turoff, M., Murray, C., de Walle, B. V., & Yao, X. (2003). The design of a dynamic emergency response management information system (DERMIS). Journal of Information Technology Theory and Application, 5(4) [article 3]. Ubeyli, E. D., & Ubeyli, M. (2009). Estimation of radiation damage at the structural materials of a hybrid reactor by probabilistic neural networks. Expert Systems with Applications, 36(3), 5184–5189. Varela, J. M., & Guedes Soares, C. (2007). A virtual environment for decision support in ship damage control. IEEE Computer Graphics and Applications, 27(4), 58–69. Zografos, K. G., Androutsopoulos, K. N., & Vasilakis, G. M. (2002). A real-time decision support system for roadway network incident response logistics. Transportation Research, 10(1), 1–18.
Contents lists available at SciVerse ScienceDirect
Expert Systems with Applications journal homepage: www.elsevier.com/locate/eswa
A knowledge-based decision support system for shipboard damage control F. Calabrese c, A. Corallo b, A. Margherita b,⇑, A.A. Zizzari a a
Centro Cultura Innovativa d’Impresa, University of Salento, Campus Ecotekne, Via Monteroni s.n., 73100 Lecce, Italy Department of Innovation Engineering, University of Salento, Campus Ecotekne, Via Monteroni s.n., 73100 Lecce, Italy c Apphia s.r.l., via Clementina Carrelli n. 26, 73100 Lecce, Italy b
a r t i c l e
i n f o
Keywords: Damage control system Decision support system Expert system Kill card Knowledge-based system Shipboard management
a b s t r a c t The operational complexity of modern ships requires the use of advanced applications, called damage control systems (DCSs), able to assist crew members in the effective handling of dangerous events and accidents. In this article we describe the development of a knowledge-based decision support system (KDSS) integrated within a DCS designed for a national navy. The KDSS uses a hybrid design and runtime knowledge model to assist damage control operators through a kill card function which supports damage identification, action scheduling and system reconfiguration. We report a fire fighting scenario as illustrative application and discuss a preliminary evaluation of benefits allowed by the system in terms of critical performance measures. Our work can support further research aimed to apply expert systems to improve shipboard security and suggest similar applications in other contexts where situational awareness and damage management are crucial. Ó 2012 Elsevier Ltd. All rights reserved.
1. Introduction Navy ships have been traditionally manned with a large crew involved in the manual control of onboard systems. The reduction of through-life costs of vessels is today a priority and there is an interest towards reducing crew without affecting damage control capabilities or jeopardizing the ability of ships to complete their missions (Cosby & Lamontagne, 2006). Beside efficiency pressures, factors like the increased complexity of modern vessels, the requirements for easy maintainability and more sophisticated operational demands generate a need for intelligent functionalities and leading-edge technologies (Bøgh & Severinsen, 2009) which can assist human decisions and actions onboard. In this endeavor, one relevant area is related to the management of events which may lead to shipboard damage and crew danger. These events require rapid actions, also without on-site human intervention, to prevent serious injuries to personnel or damages to vital ship systems. Whereas damage control has been traditionally a manual and manpower-intensive function, the automation of emergency management operations is today driven by complex technology architectures called damage control systems (DCSs) and related progresses in human-system integration, which gets increasing attention in ship design (Runnerstrom, 2003). A DCS is an information-retrieval and equipment-control system that gives ship personnel the ability to detect, analyze, and handle ⇑ Corresponding author. Address: University of Salento, Department of Innovation Engineering, c/o Euro-Mediterranean Incubator, Campus Ecotekne, Via Monteroni s.n., 73100 Lecce, Italy. Tel.: +39 0832 297922; fax: +39 0832 297927. E-mail address: [email protected] (A. Margherita). 0957-4174/$ - see front matter Ó 2012 Elsevier Ltd. All rights reserved. doi:10.1016/j.eswa.2012.01.146
various types of damage situations, based on the collection and processing of vast quantities of shipboard information. In the navy context, a damage control system (DCS) is aimed to assure the timely and informed application of men and equipment in scenarios such as fire or flooding, violation of the ship closure state, threats to essential equipments, ventilation close down, and atomic/biological/chemical issues. DCSs are also relevant for emergency training and damage instructor assistance purposes (Bulitko & Wilkins, 1999; Peters, Bratt, Clark, Pon-Barry, & Schul, 2004). The noteworthiness of these systems is proven by the number of leading market players (e.g. ABB, L3, Northrop Grumman, Rockwell, and Siemens) involved in the design and development of innovative solutions for damage control. The assistance to damage control operators, with recommendations for counteractions and reconfigurations, requires a highly structured approach to problem identification and action planning. The field of expert and decision support systems can thus provide a relevant contribution to design more performing DCSs. However, the study of expert systems and DSS in navy contexts has mostly focused on the design process whereas a very limited number of contributions have addressed the implementation of integrated systems to ensure the safety and operational stability of modern ships. In this paper, we show the development of a knowledge-based decision support system (KDSS) which has been integrated within the DCS designed for the operating needs of a national navy. We start from the analysis of the typical damage control process and identify a model of knowledge acquisition and reuse in damage management scenarios. The model is implemented through the development of a kill card function providing an interactive
F. Calabrese et al. / Expert Systems with Applications 39 (2012) 8204–8211
interface and a shared decision and action platform for damage control personnel onboard. The tool provides a graphical information-retrieval and equipment-control dashboard that gives damage crew the ability to handle various types of damage control situations. The remainder of the paper is structured as follows: Section 2 reviews existing literature on damage control and DSS applications in the navy context; Section 3 introduces the research requirements and describes the overall architectural design; Section 4 illustrates the KDSS developed; Section 5 presents an illustrative application related to a fire-fighting scenario and proposes a preliminary evaluation of quali-quantitative benefits; Section 6 concludes the paper and draws avenues for future research. 2. Damage control and DSS The concept of damage control is not limited to the navy industry. In fact, it is largely used in the field of medicine, surgery, and injury prevention as well as in project management, politics, media, and industrial production. In general, the term is used to describe the process of identifying, monitoring and dealing with any problem that may jeopardize a given system or endeavor. A first area in which information and decision support systems have been developed for damage control purposes is the field of dynamic emergency response and resource intervention (Minciardi, Sacile, & Trasforini, 2007; Turoff, Murray, de Walle, & Yao, 2003). The use of expert systems and virtual reality to support decision making in emergency management situations has been also studied (Beroggi, Waisel, & Wallace, 1995; Caro, 1992). Other contributions have investigated the use of computer-based systems to support human improvisation in extreme events (Mendonca, 2007) and the inter/intra-organizational communication and coordination in emergency response (Kanno & Furuta, 2006). As a specific area of application, the response logistics to roadway network incident (Zografos, Androutsopoulos, & Vasilakis, 2002) has been studied. Besides emergencies, damage management has represented an area of interest for expert systems and neural networks in manufacturing and engineering contexts, with a focus on structural damage detection and assessment (Barai & Pandey, 2000; Jiang, Zhang, & Zhang, 2011; Mehrjoo, Khaji, Moharrami, & Bahreininejad, 2008; Ubeyli & Ubeyli, 2009). Despite the relevance of the topic and the applicability of decision support and expert system principles, few contributions have instead focused on the development of intelligent systems for shipboard damage control. An effective damage control system (DCS) needs a systemic approach to realize key functions such as provide support to control personnel to make informed and real-time decisions, enhance total ship, coordinated, real-time control of men and equipment at the scene of damage, and allow changeover from remote-automated to local-manual control in case of emergency (Geer, 1988). In the area of intelligent applications for shipboard damage control, an expert system was created to support the cognitive processes involved in ship piloting and collision avoidance (Grabowski & Wallace, 1993). With a more specific focus on damage management, a rule-based expert system based on information from navy damage control tactics, procedures, doctrine, and experts was presented (Tate, 1996). A fuzzy distributed expert system was built to assist command and control activities (Simoes-Marques & Pires, 2003) and a virtual environment has been developed to support emergency planning decisions by considering what could occur when fluids disseminate through ship compartments, such as flooding, fire, or contamination (Varela & Guedes Soares, 2007). Finally, an expert system was developed for ship auxiliary machinery troubleshooting (Cebi, Celik, Kahraman, & Deha Er, 2009).
8205
The design and development of expert and decision support systems in ship contexts has been mostly focused on the design process (Park & Storch, 2002). Different contributions have addressed the aided design of ship systems automation (Arendt, 2004; Arendt & van Uden, 2011; Kowalski, Arendt, Meler-Kapcia, & Zielnski, 2001; Kowalski, Meler-Kapcia, Zielinski, & Drewka, 2005), the support of the conceptual design stage based on knowledge engineering (Lee, 1999; Lee & Lee, 1999), and the analysis of design problems and assessment of trade-offs between performance and cost (Chou & Benjamin, 1992). Two specific studies have developed an expert system to support compartment design of a crude oil tanker (Lee & Lee, 1997) and a DSS for vessel fleet scheduling (Fagerholt, 2004). All these studies can provide a larger background to define the requirements for ship design, building and management, with the purpose to extend the research on decision support systems applied for ship personnel security and control of casualties to ship systems. 3. Overview of the DCS The development of our KDSS was framed within a collaborative project with Avio SpA. This a leading company operating both in the aerospace propulsion (participates in military programs such as the F-35 JSF and civil partnerships with General Electric, Pratt & Whitney, and Rolls Royce) and in the marine industry for the provision of control, automation and propulsion systems and components (e.g. turbine control, lubrication and fuel systems). The company is a Marine System Supplier (MSS) of General Electric. In the last years, Avio has worked in the development of innovative damage control systems (DCSs) for national navies. The DCS is a supervisor system embedded in the Integrated Platform Management System (IPMS), which is a distributed hardware architecture used for real-time monitoring of the ship propulsion, mechanical, electrical, auxiliary and damage control systems. Monitored components include gearboxes, pitch propellers, power generation sets, power distribution switchboards, electrical distribution centers, fire pumps, systems for heating, ventilation, air conditioning, chilled water, and so forth. In practice, the IPMS controls all the onboard equipment, excluding weapons/sensors (for military ships) and the ship’s communication and navigation equipment. The general IPMS architecture comprises Multi-Function Consoles (MFCs) and Remote Terminal Units (RTUs). MFCs are mostly laptops and workstations providing the human–machine interface for the operators at various shipboard locations whereas RTUs are used for data acquisition and control and they are connected to sensors and actuators (e.g. FDS – fire detection sensors, pumps, fans). The IPMS is endowed with a runtime application allowing to monitor the whole ship from each MFC. Whereas the IPMS represents the hardware backbone for damage control operations, the DCS is the software platform configured within the IPMS with functions such as monitoring of ship subsystems, longitudinal, planar and isometric views, Tiled Layered Graphics (TLGs) approach (for automatic de-cluttering and display of complex information), casualty management, support to manage emergency states, event log and report, and compartment monitoring. These functionalities are integrated into four modules: (1) Damage Control Management System (DCMS), which enables to automatically acquire all the relevant ship safety and other data needed to handle damages, display data to the operator in an optimized way, handle alarms, and rapidly share/communicate the information between the different MFCs; (2) On Board Stability Software (OBSS), to obtain and visualize ship stability data (e.g. tanks
8206
F. Calabrese et al. / Expert Systems with Applications 39 (2012) 8204–8211
Fig. 1. IPMS hardware architecture and main DCS interface.
level, flooding sensors, water tight doors status, etc.) and calculate the stability parameters; (3) Closed Circuit TV (CCTV), which activates cameras to monitor real-time the compartments of the ship; and (4) Decision Support System (DSS), to assist the damage control officer in case of critical events by indicating the most suitable procedures to handle the specific situation. The four modules are organized within an interactive interface which allows the operator to access each function starting from a main window and navigating within the system by means of guided paths and hyperlinks. Fig. 1 shows the architecture of the IPMS and the basic interface of the DCS. The red arrow indicates that the DCS interface can be accessed from every damage control terminal (the RTU) onboard. To develop the KDSS, we first studied the decision/action process of the damage control operator onboard. We have then identified the information flows at the basis of decisions ad actions and developed the four core components of the DSS (Bonczek, Holsapple, & Whinston, 1981), i.e. the knowledge system (sources of data and information), the language system (input format), the presentation system (interface and layout) and the problemprocessing system (software engine). Next section describes the output of our work.
4. KDSS for shipboard damage control Damage and emergency management efforts onboard are coordinated by a damage control officer (DCO), which is usually supported by a damage control assistant (DCA) and a team in charge of maintaining situational awareness and taking actions to prevent injury to personnel, damage to ship systems, or loss of the ship as well. The KDSS developed has a direct impact on the decision making process and actions of these actors. A generic decision making flow includes the recognition of the problem, the listing of objectives, the perception of environment and constraints, the listing of options, the decision analysis and the action plan (Arbel & Tong, 1982). In the damage control perspective, these steps are specifically translated in a flow including five key steps (Fig. 2). The monitoring of ship systems through the damage control platform allows the operator to acquire situational awareness of damage (step one) and identify where the damage is located and which is its extent (step two). Next, the operator can start a set of preliminary actions aimed to contain and control the effects of damage (step three) and activate the damage control systems and crew to eliminate the causes of emergency and prevent further
Fig. 2. Decision and action flow of damage control operator.
F. Calabrese et al. / Expert Systems with Applications 39 (2012) 8204–8211
issues (step four). Finally, the operator reconfigures the navy and control systems after that the problem has been completely solved (step five). Throughout the five steps, the operator takes critical decisions related with what to do, what to do first, how to undertake actions, and in which order undertake actions. In order to streamline the decision flow and the execution of actions, we identified four major requirements to drive the design of the KDSS: (1) monitor all damage control status and operations at any time and from each control position onboard; (2) support damage control operations by acquiring all ship’s and security relevant data; (3) allow efficient presentations of information to the damage control operator; (4) provide decision aids, actions, procedural checklists and alarms when handling emergency situations. A KDSS is a special system for decision support which is able to recommend actions based on specialized problem-solving expertise stored as facts, rules, procedures, or in similar structures. We first designed the knowledge model underlying the system. Basic knowledge to support the decisions and actions of the damage control crew derives from two main sources: (1) design time knowledge sources, which include the ship and damage control data available at the design of the system; and (2) run time knowledge, including damage control data received ‘‘in process’’ through the damage control equipments. Design time knowledge was obtained from the ship structure and engineering data (e.g. ship layout and dimensions of compartments), navy/ship rules (e.g. operating and security management procedures) and damage control officer (e.g. engineering expertise, design suggestions, insights). Run time knowledge is implemented in the system through damage system information (e.g. fire and flooding sensors connected with the RTUs) and shared communications among the damage control operators onboard (e.g. separated actions which have to be consolidated into a unique damage checklist).
8207
Concerning the problem processing level of the KDSS, the knowledge sources are integrated within the kill card, i.e. an information and operation support dashboard which groups logically all the information related to the ship equipment and security systems, provides predefined automatic control sequences to respond to specific casualty conditions, and allows the operator to rapidly (knowing and) executing the correct actions at the moment. Since a damage generally originates in one specific compartment of the ship (and then it can propagate to other compartments), each compartment is associated to a dedicated kill card. We have thus developed a total of about 300 cards, which is the average number of compartments for a large vessel. We implemented the KDSS as a Microsoft Windows Environment developed in Visual Studio 2008. We used C# as programming language as it presents the power of C++ and the slickness of Visual Basic. Besides, the language supports the introduction of XML comments which can turn into documentation. The comments are placed into XML format and can then be used as documentation which can include example code, parameters, and references to other topics. Fig. 3 shows the overall model of our KDSS. We developed the kill card function with two main components, i.e. the editor and the viewer. The editor allows to create a new card (with a preview function), modify or delete an existing card from the kill card database if/when basic requirements related to ship structure, rules and operating procedures should change (i.e. the design time knowledge sources). The viewer allows to open and visualize an existing kill card generated through the editor. In order to retrieve a card, the operator has three options: (1) use a plan or isometric view of the ship to click on the description of a compartment; (2) choose a card from a purposeful kill card area by using a database tree structure (which describes the whole ship); (3) use a search function based on compartment names. In the viewer function, the damage control operator can also share the action list, i.e. score when response actions (e.g. in case of fire) have been started and/or completed and transmit this information to every MFC onboard. Finally, it is possible to reset changes by using a clear function and control the dynamic buttons of devices
Fig. 3. Model of the knowledge-based DSS for shipboard damage control.
8208
F. Calabrese et al. / Expert Systems with Applications 39 (2012) 8204–8211
Fig. 4. Kill card structure.
through a device window. All the information contained in every kill card and all the changes made on them are stored in a database using a XML file whose name corresponds to the ship compartment name (e.g. wardroom, engine room). The general format of a kill card (the ‘‘template’’) is an empty format structured in text fields and buttons (allowing to operate various devices) and it includes three sections or areas: (1) compartment identification area, showing critical information such as ship general data and compartment location; (2) summary area, with information related to ship and compartment sensors; (3) detail area, including several tables with information related to ventilation control, danger hazards, fire fighting installations, actions, compartment views and video clips, and status of devices. The damage control operator can visualize actions from a predefined list and ‘‘check off’’ (see Fig. 4) when these actions have been successfully completed. The system also allows to link kill cards of different compartments when the effects of damage propagate into different ship rooms. In this way, it is possible to use the sensors and devices available in more compartments, identify priority areas, execute shared actions, and so forth. The basic structure and content of a kill card template can be modified through a purposeful configurator. The editor, viewer and configurator components are operationalized by a software application available in every MFC of the ship, allowing damage
control operators to be always in possess of the information pertinent to the emergency situation managed at the moment. The kill card ‘‘uses’’ the design information (such as compartments layout and dimensions, ship security management procedures, and damage control officer insights) and run time knowledge (like system status, status of actuators and shared information on actions taken by different operators), to provide expert assistance to the damage control crew. The kill card is the main element of the user interface and it allows a dynamic situational awareness of damage, damage localization and extent, preliminary counteractions, damage control systems and crew activation and reconfiguration. The KDSS suggests therefore to the damage control operator what to do, what to do first, how to undertake actions, and in which order undertake actions in the different emergency situations which may happen onboard. Next section shows an illustrative scenario of fire fighting in which the system can provide its support to damage control operations. A preliminary evaluation of benefits achieved is also described.
5. Illustrative scenario and evaluation of benefits We describe hereafter an illustrative scenario in which a fire emergency onboard requires rapid actions to avoid serious injuries to personnel or damages to vital ship systems. In this case, the damage control crew is involved in the following sequence of steps:
F. Calabrese et al. / Expert Systems with Applications 39 (2012) 8204–8211
(1) acquire awareness about a fire happened onboard; (2) identify where fire is located and which is its extent; (3) start a set of preliminary actions aimed to contain and control the effects of fire; (4) activate damage control systems and crew to eliminate causes of fire and prevent further damages; (5) reconfigure and restore the ship and damage control systems after fire. The five steps are supported by the integrated functionalities allowed by the IPMS and the DCS of the ship. In particular, the KDSS developed has a direct impact on steps 3 and 4. In the first step, the sensors installed in the compartment concerned with the fire transmit an alarm or warning message which is visible in the summary area of the kill card interface. The operator can thus immediately become aware of the emergency. In step two, the system allows the operator to visualize which compartment is concerned and which is the seriousness of damage. These analyses are supported by different visualization options of the compartment and the use of a zoom feature and red signs on the interface (in Fig. 5 a red box in the ship layout indicates the compartment where fire has started). Besides, devices and sensors acquire quantitative measures (e.g. temperature, pressure, CO2 level, etc.) which are sent to the system for real-time damage evaluation. After that, the operator can start a set of preliminary actions (step 3) aimed to contain and control the effects of fire. At this purpose, the operator access the kill card database to retrieve the specific kill card of the compartment concerned with the fire. In fact, this kill card will indicate the list of actions suggested by the KDSS (on the basis of the knowledge model underlying the system) to resolve the problem. An example of action list in case of fire is the following: (1) close ventilation; (2) preserve watertight integrity; (3) maintain vital systems; (4) isolate, prevent, extinguish, combat and remove the effects of fire; (5) facilitate the care of personnel casualties; and (6) make rapid repairs to the ship’s structure and equipment. The operator can then activate the damage control systems and crew (step 4) to prevent further damage and limit current issues.
8209
For example, the operator can monitor the status of devices and control remotely pumps and fans which are present at the fire scene. Then, he/she can assemble the damage control task force and execute the secondary action plan with the purpose to completely extinguish the fire. When the alarm status is over, the operator can then reconfigure and restore the ship and damage control systems. Through the remote control functionalities of the KDSS, the sensors and devices, as well as all the action lists turned on can be reset and made ready in case of a new emergency. The fire fighting case, along with other damage management scenarios, was created to test the behavior of the crew and the system through the damage management process. To validate the KDSS, we used indeed a combination of technical, empirical and subjective methods (Adelman, 1992; Borenstein, 1998; Papamichail & French, 2005). We realized a technical validation by cross-checking the information retrieval and visualization output obtained from different damage control stations onboard, with a specific attention to the consistency of data obtained and knowledge sources used. In particular, we ascertained that all the MFCs on board show the same data to the different damage control operators. We obtained empirical validation through simulations aimed to verify the correct functioning of the tool and measure its performance as enhancer of the decision making and action process. A combination of outcome and process criteria is in fact highly relevant for DSS evaluation (Phillips-Wren, Hahn, & Forgionne, 2004). Empirical validation was strictly linked with the subjective assessment obtained with user confirmation about the benefits/utility and usability of the tool. At this purpose, we interviewed the ship damage officer and operators to understand if the system meets the needs of users and how well the interface was designed. We have also measured the benefits allowed by the DSS on the basis of four critical metrics and four qualitative indicators. Metrics mostly refer to the impact in terms of user and damage control process performance and include: (1) awareness time, i.e. the average time needed by the damage control operator to acquire situation awareness after that a damage event happens; (2) action time,
Fig. 5. Kill card windows in the fire fighting scenario.
8210
F. Calabrese et al. / Expert Systems with Applications 39 (2012) 8204–8211
Table 1 Indicators and benefits allowed by the KDSS. Indicator
Without KDSS
With KDSS
Awareness time Action time Crew need Action cost Navigability Multimediality Information sharing Interoperability
15 min 45 min 5 Operators 112,5$ Not allowed Not allowed Docs, phone, meetings
1 min 15 min 2 Operators 15$ Hypertexts and function trees Audio/video MFCs
Not allowed
Integrated DCS
i.e. the average time needed to activate the key procedures to solve the problem or reduce the damage; (3) crew need, i.e. the average number of total crew members required to act throughout the whole damage handling process; (4) action cost, i.e. the average cost for the complete end-to-end decision and action procedure. Qualitative indicators refer to system features in terms of: (1) navigability, i.e. the direct and fast access to single functions of the tool; (2) multimediality, i.e. the possibility to integrate different forms of information visualization and human–system interaction tools; (3) information sharing, i.e. the availability of the same damage information and actions status at the different control positions onboard; and (4) interoperability, i.e. the integration of different software applications with effects which are immediately implemented within the different tools. Table 1 reports a comparison of indicators in the ‘‘without KDSS’’ and ‘‘with KDSS’’ scenarios. ‘‘Without KDSS’’ values are average measures reported by navy personnel involved in damage control operations whereas after-implementation values (‘‘with KDSS’’) derive from a preliminary estimation based on simulations and managerial/operator validation at the utilization site. The situational awareness time is about 15 min without the system. In fact, a visible damage or problem in a ship system or compartment activates the internal crew communication process through which the warning message is received by the damage control operator. The use of kill cards connected with sensors and cameras allows to reduce such time to few seconds (and less than 1 min) needed by the operator to visualize the automatic warning message on the screen (with an improvement of about 93%). The damage action time can be decreased from 45 to about 15 min (improvement of 67%), as the operator can now immediately start a set of corrective actions directly from the remote terminal unit and there is no need to be physically present at the damage site. The KDSS provides the operator with a kill card which synthesizes all the information and knowledge critical to successfully handle the problem. Naturally, damage action time represents a very critical measure for the safety of personnel and the survival of vital ship systems. Concerning crew need, five persons are needed on average in the damage control staff to handle the situation from end to end. In the with-KDSS scenario, only two persons (60% decrease) are required to use the interface and support the overall damage recognition and recovery action process. As a consequence of reduction in action time and crew need, the total action cost related with the employment of specialized crew (materials, resources consumption and other costs are thus left out in this calculation) can be now reduced from 112,5$ to about 15$, with a reduction of about 87% (we made an assumption of 30$ for the average hour cost of a damage crew member). Moreover, the automation of routine actions allows to further improve the performance of damage control operations since the overall efficiency of operations is increased and damage personnel can be involved in value added activities connected with damage control or other tasks related with shipboard management.
Concerning qualitative benefits, navigability, multimediality, information sharing and interoperability are new features allowed by the KDSS and the integrated damage control platform. Navigability of the tool is supported by the use of hyperlinks and function navigation trees. The system allows some multimedia features with audio and video signals directly coming from ship compartments. Information sharing, which was based on exchange of documents, physical meetings and phone calls, is now supported by the direct communication among the Multi-Function Consoles (MFCs) on board. Finally, the system also supports interoperability as the IPMS, the modules of the DCS and the KDSS are fully integrated and can be customized based on user needs and requirements.
6. Conclusions The identification and management of events that may lead to shipboard damage and crew danger are interesting areas of application for expert and decision support methods and tools. National navies around the world are in fact turning to enhanced and distributed damage control systems to achieve higher level of security and operational efficiency through effective information sharing, fast problem identification and action planning and automation. In this paper, we have presented a knowledge based DSS which uses design time and run time knowledge sources to streamline the decision making process and sequence of actions required to the damage control operator in case of emergency. The system reduces situational awareness time, action time, crew need and overall action cost. The KDSS also allows full navigability of damage control information, the use of multimedia tools for damage monitoring, the interoperability with other DCS applications, and a more effective information sharing. The dedicated displays onboard enable the operators to immediately identify the emergency and initiate corrective actions. The ship-wide data network allows several dispersed damage stations to retrieve coherent information and thereby effectuate a coordinated and effective action, resulting in reduced damage control response time, enhanced consistency of actions, and reduced manning. The use of the application developed could be enlarged to other contexts (e.g. building sites, nuclear and other energy production sites) where the monitoring of risky events in different compartments or operation areas requires advanced control and decision/ action support technologies. In such cases, the KDSS can be of value to increase the situational awareness of damage crew members and enhance data consistency through the use of automatized control devices for the remote identification of risks. Next research will be addressed to extend the application of the DSS for training purposes, and in particular for on-the-job training of damage crew members. The onboard training system of the ship could be indeed used to simulate events in normal ship operation as well as in degraded conditions using the same interface. A second area of development is represented by the adoption of enhanced reality and 3D technologies and functionalities within the system. This could further enhance the situational awareness of operators and their ability to promptly identify and assess the problem, resulting in faster and more effective actions.
Acknowledgments The authors are grateful to the persons at Avio SpA who have collaborated in the design and implementation activities. A particular acknowledgment goes to Marco Rosso for his support in the final revision of the article.
F. Calabrese et al. / Expert Systems with Applications 39 (2012) 8204–8211
References Adelman, L. (1992). Evaluating decision support and expert systems. New York, NY: Wiley. Arbel, A., & Tong, R. (1982). On the generation of alternatives in decision analysis problems. Journal of the Operational Research Society, 33, 377–387. Arendt, R. (2004). The application of an expert system for simulation investigations in the aided design of ship power systems automation. Expert Systems with Applications, 27(3), 493–499. Arendt, R., & van Uden, E. (2011). A decision-making module for aiding ship system automation design: A knowledge-based approach. Expert Systems with Applications, 38(1), 410–416. Barai, S. V., & Pandey, P. C. (2000). Integration of damage assessment paradigms of steel bridges on a blackboard architecture. Expert Systems with Applications, 19(3), 193–207. Beroggi, G. E. G., Waisel, L., & Wallace, W. A. (1995). Employing virtual reality technology to support decision making in emergency management. Safety Science, 20(1), 79–88. Bøgh, S. A., & Severinsen, T. (2009). Damage management control system on the Danish navy ships. In Proceedings of the 14th international ship control systems symposium (SCSS). Ottawa, Canada. Bonczek, R. H., Holsapple, C., & Whinston, A. B. (1981). Foundations of decision support systems. New York, NY: Academic Press. Borenstein, D. (1998). Towards a practical method to validate decision support systems. Decision Support Systems, 23(3), 227–239. Bulitko, V. V. & Wilkins, D. C. (1999). Automated instructor assistant for ship damage control. In Proceedings of the 16th national conference on artificial intelligence. Orlando, FL: American Association for Artificial Intelligence. Caro, D. H. J. (1992). Expert support systems in an emergency management environment – An evaluative framework. Educational and Training Technology International, 29(2), 132–142. Cebi, S., Celik, M., Kahraman, C., & Deha Er, I. (2009). An expert system towards solving ship auxiliary machinery troubleshooting: SHIPAMTSOLVER. Expert Systems with Applications, 36(3), 7219–7227. Chou, Y.-C., & Benjamin, C. O. (1992). AI-based decisions support system for naval ship design. Naval Engineers Journal, 104(3), 156–164. Cosby, L., & Lamontagne, Y. (2006). Critical assessment of damage control system technologies. Defence R&D Canada Report. Fagerholt, K. (2004). A computer-based decision support system for vessel fleet scheduling – Experience and future research. Decision Support Systems, 37(1), 35–47. Geer, D. (1988). Advanced damage control systems. Naval Engineers Journal, 100(3), 143–153. Grabowski, M., & Wallace, W. A. (1993). An expert system for maritime pilots: Its design and assessment using gaming. Management Science, 39(12), 1506–1520. Jiang, S.-F., Zhang, C.-M., & Zhang, S. (2011). Two-stage structural damage detection using fuzzy neural networks and data fusion techniques. Expert Systems with Applications, 38(1), 511–519. Kanno, T., & Furuta, K. (2006). Modeling and simulation of inter-and intraorganizational communication and coordination in emergency response. International Journal of Emergency Management, 3(2–3), 149–167.
8211
Kowalski, Z., Arendt, R., Meler-Kapcia, M., & Zielnski, S. (2001). An expert system for aided design of ship systems automation. Expert Systems with Applications, 20(3), 261–266. Kowalski, Z., Meler-Kapcia, M., Zielinski, S., & Drewka, M. (2005). CBR methodology application in an expert system for aided design ship’s engine room automation. Expert Systems with Applications, 29(2), 256–263. Lee, D.-K. (1999). Hybrid system approach to optimum design of a ship. Artificial Intelligence for Engineering Design, Analysis and Manufacturing: AIEDAM, 13(1), 1–11. Lee, K.-H., & Lee, K.-Y. (1997). Knowledge-based nonmonotonic reasoning process in ship compartment design system. Expert Systems with Applications, 13(2), 145–154. Lee, D., & Lee, K.-H. (1999). An approach to case-based system for conceptual ship design assistant. Expert Systems with Applications, 16(2), 97–104. Mehrjoo, M., Khaji, N., Moharrami, H., & Bahreininejad, A. (2008). Damage detection of truss bridge joints using artificial neural networks. Expert Systems with Applications, 35(3), 1122–1131. Mendonca, D. (2007). Decision support for improvisation in response to extreme events: Learning from the response to the 2001 world trade center attack. Decision Support Systems, 43(3), 952–967. Minciardi, R., Sacile, R., & Trasforini, R. (2007). A decision support system for resource intervention in real-time emergency management. International Journal of Emergency Management, 4(1), 59–71. Papamichail, K. N., & French, S. (2005). Design and evaluation of an intelligent decision support system for nuclear emergencies. Decision Support Systems, 41(1), 84–111. Park, J.-H., & Storch, R. L. (2002). Overview of ship-design expert systems. Expert Systems, 19(3), 136–141. Peters, S., Bratt, E. O., Clark, B., Pon-Barry, H., Schul, K. (2004). Intelligent systems for training damage control assistants. In Interservice/industry training, simulation, and education conference (I/ITSEC). Phillips-Wren, G. E., Hahn, E. D., & Forgionne, G. A. (2004). A multiple-criteria framework for evaluation of decision support systems. Omega-International Journal of Management Science, 32(4), 323–332. Runnerstrom, E. (2003). Human systems integration and shipboard damage control. Naval Engineers Journal, 115(4), 71–80. Simoes-Marques, M. J., & Pires, F. J. (2003). SINGRAR – A fuzzy distributed expert system to assist command and control activities in naval environment. European Journal of Operational Research, 145(2), 343–362. Tate, D. L. (1996). Development of a tactical decision aid for shipboard damage control. NRL, Report NRL/FR/5580-96-9837. Turoff, M., Murray, C., de Walle, B. V., & Yao, X. (2003). The design of a dynamic emergency response management information system (DERMIS). Journal of Information Technology Theory and Application, 5(4) [article 3]. Ubeyli, E. D., & Ubeyli, M. (2009). Estimation of radiation damage at the structural materials of a hybrid reactor by probabilistic neural networks. Expert Systems with Applications, 36(3), 5184–5189. Varela, J. M., & Guedes Soares, C. (2007). A virtual environment for decision support in ship damage control. IEEE Computer Graphics and Applications, 27(4), 58–69. Zografos, K. G., Androutsopoulos, K. N., & Vasilakis, G. M. (2002). A real-time decision support system for roadway network incident response logistics. Transportation Research, 10(1), 1–18.