- Email: [email protected]
Analyzing different material supply strategies in matrix-structured manufacturing systems
Available online at www.sciencedirect.com Available online at www.sciencedirect.com
ScienceDirect ScienceDirect Available online atonline www.sciencedirect.com Procedia CIRP 00 (2019) 000–000 Available at www.sciencedirect.com
ScienceDirect ScienceDirect
Procedia CIRP 00 (2019) 000–000
www.elsevier.com/locate/procedia www.elsevier.com/locate/procedia
ProcediaProcedia CIRP 00CIRP (2017) 81000–000 (2019) 1004–1009
52nd CIRP Conference on Manufacturing Systems 52nd CIRP Conference on Manufacturing Systems
www.elsevier.com/locate/procedia
Analyzing different material supply strategies in matrix-structured Analyzing different supplyMay strategies in matrix-structured 28th CIRPmaterial Design Conference, 2018, Nantes, France manufacturing systems manufacturing systems a a A newMarc-André methodology analyze the functional and physical architecture of Filza,*,to Johann Gerberding , Christoph Herrmann , Sebastian Thiedea a,* a a Marc-André FilzTools , Johann Gerberding , Christoph Herrmannfamily , Life Sebastian Thiedea existing products for anProduction assembly product Institute of Machine and Technology, oriented Chair of Sustainable Manufacturing and Cycleidentification Engineering, a
Technische Universität Braunschweig, Langer Kamp 19 b, 38106 Braunschweig, Germany Institute of Machine Tools and Production Technology, Chair of Sustainable Manufacturing and Life Cycle Engineering, * Corresponding author. Tel+49-531-390-65032; fax: +49-531-391-5842. Technische Universität Braunschweig,E-mail Langeraddress: Kamp [email protected] b, 38106 Braunschweig, Germany a
Paul Stief *, Jean-Yves Dantan, Alain Etienne, Ali Siadat
* Corresponding author. Tel+49-531-390-65032; fax: +49-531-391-5842. E-mail address: [email protected] École Nationale Supérieure d’Arts et Métiers, Arts et Métiers ParisTech, LCFC EA 4495, 4 Rue Augustin Fresnel, Metz 57078, France
Abstract
* Corresponding author. Tel.: +33 3 87 37 54 30; E-mail address: [email protected]
Abstract The matrix-structured manufacturing system represents an alternative manufacturing concept that strives for both a volume and variant flexible production while still being efficient. The manufacturing system is a modular and cycle-independent production principle in which all existing The matrix-structured manufacturing system represents an alternative manufacturing concept that strives for both a volume and variant flexible workstations are linked together. By using intelligent decentralized control logic, the assignment of workpieces to workstations is enabled. Due Abstract production while still being efficient. The manufacturing system is a modular and cycle-independent production principle in which all existing to the functional principle of a matrix-structured manufacturing system, the material requirements at the individual workstations are not known workstations are linked together. By using intelligent decentralized control logic, the assignment of workpieces to workstations is enabled. Due advance. The behavior of different material supply strategies can be investigated with theishelp of an agent-based simulation. In this context, Inin business environment, the trend towards more productsystem, variety andmaterial customization unbroken. Due to thisworkstations development, of totoday’s the functional principle of a matrix-structured manufacturing the requirements at the individual arethe notneed known the objective of this paper is to model and analyze selected material supply strategies with regard to selected key figures in an agent-based agile and reconfigurable production systems emerged to cope with various products and product families. To design and optimize production in advance. The behavior of different material supply strategies can be investigated with the help of an agent-based simulation. In this context, simulation environment. systems as well to paper chooseisthe optimaland product matches, product analysis arewith needed. Indeed, most of knowninmethods aim to the objective ofasthis to model analyze selected material supplymethods strategies regard to selected keythefigures an agent-based analyze a product or one product family on the physical level. Different product families, however, may differ largely in terms of the number and simulation environment. © 2019 Authors. Published Elsevier This iscomparison an open access the CC BY-NC-ND license nature ofThe components. This fact by impedes anLtd. efficient and article choice under of appropriate product family combinations for the production © 2019 The Authors. Published by Elsevier Ltd. (http://creativecommons.org/licenses/by-nc-nd/3.0/) system. AThe new methodology is proposed to analyze existing products in article view ofunder theirthe functional and physical architecture. The aim is to cluster © 2019 Authors. Published by Elsevier Ltd. This is an open access CC BY-NC-ND license This is an open access article under the scientific CC BY-NC-ND license (http://creativecommons.org/licenses/by-nc-nd/3.0/) Peer-review responsibility of the committee of the 52nd CIRPofConference on Manufacturing Systems. these productsunder in new assembly oriented product families for the optimization existing assembly lines and the creation of future reconfigurable (http://creativecommons.org/licenses/by-nc-nd/3.0/) Peer-review under responsibility of the scientific committee of the 52nd CIRP Conference on Manufacturing Systems. assembly systems. on Datum Chain, the physicalofstructure the products is analyzed. FunctionalSystems. subassemblies are identified, and Peer-review underBased responsibility of Flow the scientific committee the 52ndofCIRP Conference on Manufacturing manufacturing system; Logistic concepts; Manufacturing simulation; Intelligent manufacturing systems; Flexibility a Keywords: functionalMatrix-structured analysis is performed. Moreover, a hybrid functional and physical architecture graph (HyFPAG) is the output which depicts the similarity product manufacturing families by providing design supportManufacturing to both, production planners and product designers. Keywords:between Matrix-structured system; Logistic concepts; simulation;system Intelligent manufacturing systems; FlexibilityAn illustrative example of a nail-clipper is used to explain the proposed methodology. An industrial case study on two product families of steering columns of thyssenkrupp Presta France is then carried out to give a first industrial evaluation of the proposed approach. 1. Introduction components and thus has a great influence on the efficiency of © 2017 The Authors. Published by Elsevier B.V. theDesign wholeConference production system. network structure of the Peer-review under responsibility of the scientific committee of the 28th CIRP 1. Introduction components and thus2018. has a greatThe influence on the efficiency of
In numerous industrial sectors, assembly line production is
MMS, which consists ofsystem. an uncoupled modularstructure and redundant the whole production The network of the WS design, results in multiple demand locations forredundant the same MMS, which consists of an uncoupled modular and currently exposed toprinciple. an increasing amount of areas variants, material or results modules. As a result, thelocations destinations the the main production Companies in these are WS design, in multiple demand for theofsame decreasing product life cycles and a strongly fluctuating materials are not known in advance. The production control currently exposed to an increasing amount of variants, material or modules. As a result, the destinations of the demand [1].product This results in corresponding flexibility and system assigns theknown products to the WS atproduction short notice and decreasing life cycles and a strongly fluctuating materials are not in advance. The control 1.efficiency Introduction of the product range and characteristics manufactured and/or requirements for the production principle. The depends on the particular circumstances, which limits the demand [1]. This results in corresponding flexibility and assembled system assigns the products tocontext, the WSthe at main shortchallenge notice and in this system. In this in recently developed concept of a matrix manufacturing system ability to plan the provision and increases the requirements for efficiency requirements for the production principle. The depends on the particular circumstances, which limits the Due to the fast development in the domain of modelling and analysis is now not only to cope with single provides developed the necessary high operational flexibility and flexibility and the responsiveness. recently of a matrix ability to aplan provision and increases theproduct requirements for communication andconcept an ongoing trend manufacturing of digitizationsystem and products, limited product as range families, scalability [2]. Due to the complexity wellorasexisting dynamic and stochastic provides the necessary high operational flexibility and flexibility and responsiveness. digitalization, manufacturing manufacturing enterprises are facing also to bethis ablemanufacturing to analyze andapproach, to compare to define The matrix-structured systemimportant (MMS) but behavior theproducts requirements for scalability [2]. Due toofthe complexity as well as dynamic and stochastic challenges in today’s market environments: a continuing new product families. It can be observed that classical existing consists of modular uncoupled workstations (WS) which are production logistics increase. To overcome these challenges, The matrix-structured manufacturing system (MMS) behavior of this manufacturing approach, the requirements for tendency reduction of product development times families are regrouped in function of clients or features. connectedtowards a flexible transportation system (WS) (see Fig. 1).and An product this paper develops simulation for different material consists ofby modular uncoupled workstations which are production logistics aincrease. To model overcome these challenges, shortened product lifecycles. In addition, there is an increasing However, assembly oriented product families are hardly to find. intelligent production control system coordinates the material supply strategies within the MMS with regard to selected key connected by a flexible transportation system (see Fig. 1). An this paper develops a simulation model for different material demand of customization, being at the same time in a global On the product family level, products differ mainly in two flow of the products through the production system. An performance indicators. intelligent production control system coordinates the material supply strategies within the MMS with regard to selected key competition with competitors all over the world. This trend, main characteristics: (i) the number of components and (ii) the elementary of every is the flow of thecomponent products through theproduction productionsystem system. An performance indicators. which the development from macro micro type of components (e.g. mechanical, electrical, electronical). internalis inducing material supply, whichproduction provides the torequired elementary component of every system is the markets, results in diminished lot sizes due to augmenting Classical methodologies considering mainly single products internal material supply, which provides the required product varieties (high-volume to low-volume production) [1]. or solitary, already existing product families analyze the 2212-8271 © 2019 The Authors. Published by Elsevier Ltd. This is an open access article under the CC BY-NC-ND license To cope with this augmenting variety as well as to be able to product structure on a physical level (components level) which (http://creativecommons.org/licenses/by-nc-nd/3.0/) 2212-8271 © 2019 The Authors. Published by Elsevier Ltd. This is an open access article under the CC BY-NC-ND license an efficient definition and identify possible optimization potentials in the existing causes regarding Peer-review under responsibility of the scientific committee of the 52nd CIRP Conference on difficulties Manufacturing Systems. (http://creativecommons.org/licenses/by-nc-nd/3.0/) production system, it is important to have a precise knowledge comparison of different product families. Addressing this theInmain production principle. Companies in areas are Keywords: Assembly; Design method; Family identification numerous industrial sectors, assembly linethese production is
Peer-review under responsibility of the scientific committee of the 52nd CIRP Conference on Manufacturing Systems.
2212-8271 © 2019 The Authors. Published by Elsevier Ltd. This is an©open article Published under theby CC BY-NC-ND 2212-8271 2017access The Authors. Elsevier B.V. license (http://creativecommons.org/licenses/by-nc-nd/3.0/) Peer-review under responsibility of scientific the scientific committee theCIRP 52ndDesign CIRPConference Conference2018. on Manufacturing Systems. Peer-review under responsibility of the committee of the of 28th 10.1016/j.procir.2019.03.242
2
Marc-André Filz et al. / Procedia CIRP 81 (2019) 1004–1009 Marc-André Filz et al. / Procedia CIRP 00 (2019) 000–000
2. Material supply strategies for matrix manufacturing systems 2.1. Requirements for material supply strategies In order to be able to define the requirements for material supply in MMS, the influencing factors on material supply and its design must be determined. Since material supply is a logistics system, it is part of the superordinate production system. Characteristics of the production system thus define the framework conditions for the material supply. In addition, the logistical (e.g. usage frequency), physical (e.g. volumes) and handling (e.g. bulk goods) characteristics of the material spectrum to be provided have a large influence on the design of the material supply [3]. In order to gain a better understanding of the requirements of MMS, a comparison of a line and MMS configuration is shown in Fig. 1.
1005
Table 1. Comparison of requirements regarding line and MMS configuration. Requirements
Line configuration
MMS configuration
Delivery locations
single dependant on assembly order
multiple
Material summary Short-term capacity planning
fix
hardly possible variable/ dynamic
In addition, due to lot size 1, the amount of transport operations within the MMS increases. Due to the network structure of the system, there is an overlap of the main material flow and the production supply flow. In combination with a high number of transport operations, this increases the risk of blockages on the transport routes and makes the overall system very complex. So-called deadlocks can occur especially in close-meshed layouts and represent critical situations. For example, decentrally controlled automated guided vehicles (AGV) trigger a circular closing and completely block each other and need to be prevented [4]. 2.2. Material supply strategies for MMS
Fig. 1. Material supply strategies for MMS.
The MMS network structure of unlinked modular working stations and its multiple redundant design result in multiple delivery locations for the same material or modules. The layout of the individual WS must be designed in a way that the scope of supply of the respective WS can be easily arranged. In addition, source-sink relationships in an MMS are variable and behave dynamically depending on the respective system status and the selected production control logic. Therefore, different supply locations within the production system exist for each workpiece (WP). A summary of material is thus hardly possible due to a missing planning basis. After the material infeed into the production system, the WS on which the product is processed are unknown as they are determined at short notice and depending on the situation by the production control system. This results in a missing detailed short-term capacity planning. Table 1 gives an overview of the different requirements regarding the line and MMS configuration.
The initial basis for the development of material supply strategies for MMS is a logistical segmentation regarding the article spectrum to be provided. The design of the material staging process depends largely on the application context and usually consists of a combination of different concepts, each of which is applied to a specific segment of the article spectrum. For the supply of components in an MMS, which are characterized by small batch sizes as well as a high variety of types and variants, product-specific order supply as a set or kit is recommended. Parts and modules that can be combined as a set with respect to their volumes should be delivered to the WS pre-commissioned in order to keep inventory levels within the system as low as possible. Due to the uncertainties with regard to the production schedule of a MMS, the set and module commissioning per product occurs in a supermarket close to the production. Individual supply as direct delivery is suggested in the case of large-volume parts which cannot be supplied within a set due to their volumes or geometric properties. This ensures optimum use of space and transparency within the production environment. In this context, it is necessary to consider that a just-in-sequence delivery (JIS) is not possible due to the lack of a stable order sequence in the MMS. This paper focuses on material transport between supermarket and WS within the MMS. This results in three possible material supply strategies for MMS that are shown in Fig. 2.
Fig. 2. Selected material supply strategies for MMS.
A supply by shopping basket is a form of material supply (e.g. JIT) and is frequently used in the automotive industry. The
1006
Marc-André Filz et al. / Procedia CIRP 81 (2019) 1004–1009 Marc-André Filz et al. / Procedia CIRP 00 (2019) 000–000
parts within the shopping basket are optimally arranged ergonomically according to their respective assembly order. Depending on the capacity of the shopping basket, it can provide the required material for one or more WP of a product. This material supply is particularly suitable for small and medium-sized parts with many variants [5,6]. The internal transport of the shopping basket can be carried out within the MMS by the product specific, automated and guided vehicle system. If the capacity of a shopping basket is sufficient for the number of parts to be assembled, no additional logistical effort is required. If the capacity is insufficient, it must be changed or refilled during the production process. In order to avoid longer interruptions in the production process and long transport distances, logistics stations can be integrated within the MMS. If an attached shopping basket has been exhausted, the product AGV drives to the logistics station and the attached empty shopping basket is replaced by a new shopping basket. Afterwards the product AGV drives to the next WS. The supply of the parts by tugger train system is an alternative to the shopping basket supply, which also has no restrictions regarding the part size. After a WS has been supplied with the required materials, empty containers are also returned. The number of routes and the routing within an MMS are variable to ensure that the material supply can adapt to the respective conditions and react flexibly. In addition, the performance of a tugger train system depends on the layout design, in particular the positioning of the material supply source. In the MMS, this source represents a supermarket which can be positioned within or outside the manufacturing system. An automated guided vehicle system for material supply in MMS is a highly complex system that transports the entire material spectrum of a production program. Compared to the supply by shopping basket, the AGV has no fixed allocation to a specific product. It is therefore necessary to combine different AGV in a complete logistics system. The structure of a MMS results in a complex network of different transport routes. The AGV can operate autonomously within the network for the supply and disposal of the production. Within the AGV, the selection of the vehicle depends on the materials to be transported [7]. 2.3. Simulation of matrix-structured manufacturing systems In industry and science, simulation is a well-known methodology for representing or imitating real system over time [8]. Since production logistics systems are usually quite complex, simulation is often used for the validation, analysis and optimization of the systems [9]. In the case of manufacturing, a simulation method is used as a supporting method for several tasks. Often it is used for layout design, planning, analysis and optimization of manufacturing systems [10]. Furthermore, a simulation of production can help to analyse a system with regard to its cause-effect relationships and to make the system behavior visible to the observer [8,10]. A widely used method for production simulation is discrete event simulation [10]. Therefore, passive entities, such as people or tasks, move through a manufacturing system and trigger actions at discrete events over time. Moreover, another paradigm is the approach of agent-based simulation. This
3
approach focuses on decentralized modelling of the behavior of individual entities within a predefined environment [11]. Each element within this system acts according to its own individual logic and interacts dynamically with other elements. In addition, agent-based simulation approaches can be used to model, simulate and control manufacturing systems [11–13]. Accordingly, an appropriate simulation model for the analysis of the described logistics concepts is also needed within the scope of this paper. Regarding the applications for MMS, [2] present a methodology that enables assembly line production to combine high flexibility with profitability due to the elimination of identical cycle times while maintaining a seamless process. The focus of this paper is on a systematic assignment of several operating steps to specially equipped WS and a control system that controls the corresponding distribution and ensures the dynamic configurability of the system [2]. Moreover, [14] discuss the main principles, elements and control strategies of a MMS. In addition, the paper introduces a simulation approach for the evaluation of the MMS control strategies. However, the simulation approach is applied to a use case and measures are derived on how this approach can be used for the planning of MMS. A detailed consideration of logistics is not the primary focus of the work. [15] focus on an agent-based simulation approach for industrial grade software tools to increase the flexibility of the manufacturing system. Therefore, a general methodology for implementing agent-based logic is presented based on a MMS use-case. With more focus on logistics, [16] describe an integrated logistics concept for the future modular final assembly of automobiles, consisting of five different material supply strategies. In the course of a case study in pre-assembly at a German automobile manufacturer, a space reduction of 15 % was achieved with the help of this concept. However, there is no validation or consideration of the performance of the concept. Moreover, [17] describe a hybrid transport concept for the material supply of a modular production system using tugger trains in combination with automated guided vehicles. A validation of the presented concept is not available in this context. However, none of these approaches regarding the simulation of MMS and logistics concept anticipate operational influences, like utilization or efficiency, for logistic planning and optimization. Therefore, no conclusions about the validity or operative effects of the material supply strategy on the system can be made. 3. Simulation model for the analysis of material supply concepts in matrix-structured manufacturing systems The previous investigations regarding the efficiency of the MMS assume that the production logistics is able to provide the required material at the right time at the right place in the right quality. However, due to the fact that the MMS is a highly dynamic and hardly predictable system, detailed planning of the material supply is not possible. As a result, the requirements placed on production logistics in terms of flexibility and responsiveness are increasing significantly. Therefore, a model is presented in the following, which examines the material
4
Marc-André Filz et al. / Procedia CIRP 81 (2019) 1004–1009 Marc-André Filz et al. / Procedia CIRP 00 (2019) 000–000
supply concepts described previously for their performance in a MMS. 3.1. Development of simulation model architecture The MMS simulation model within this paper consists of nine WS on which different products are manufactured. Each of the WS is able to carry out different WP and consists of a processing workplace and a buffer space with the capacity of one product (see Fig. 1). Since the production processes in a MMS can, for example, also involve assembly activities, which are usually carried out by employees, the processing times of the WP are not constant. These can vary from employee to employee for various reasons (e.g. age, qualification, learning curve), which is why the processing times are assumed to be normally distributed within the scope of this work. The material flow of the products is carried out by the selected material supply strategy (see Fig. 2). The routing to the respective successor station is determined by the product itself. It queries the status data of all possible successor stations and calculates the corresponding transport times. In accordance with [14] the WS is selected at which the time to start processing the product is the shortest. The maximum number of products within the MMS is limited to prevent blockades of the entire system [14]. Moreover, this maximum number of products within the system depends on the system state. With the help of a simulation, the optimal maximum number of products simultaneously within the MMS was determined as nine products. Fig. 3 shows the main object classes of the developed simulation model and some of the most important class parameters, variables and functions.
Fig. 3. Simplified representation of model classes.
The basic MMS concept considered within this paper is based on [14]. In addition to the products and WS, the material provision in the form of two agents for the transport system and the transport orders is integrated within this simulation architecture. The product agents are able to navigate autonomously through the production system according to the production control settings. Moreover, the product agents
1007
trigger the material supply for their next WP themselves by running an (OrderMaterial()) function after completing the previous work package. The corresponding material requirements are autonomously assigned to each WP. The order function creates transfer orders for the required material, which contain the staging location and quantity. Furthermore, it transfers them to an order list. The main parameters of the transporter agent are the type of transportation and loadable capacity. The specific behavior of the agent is modelled by a state diagram shown in Fig. 4.
Fig. 4. Statechart of transport agent.
Within this architecture, it is assumed that the material provision system ensures 100 % availability. At the start of the simulation, the transport vehicles are all waiting in their home positions. A search function (SearchOrder()) is called every five seconds to check a list of transport requests. If one or more requests exist, the transporter selects the first request from the list and executes it. In the first step, it drives to the loading station and stays there for the corresponding loading time period. Subsequently, the transfer to the staging point occurs where a certain amount of time is spent waiting during unloading before returning to the home position and selecting the next transport order. The routing of the transport vehicles takes place automatically using the shortest route within the network to the provisioning point, whereby collisions and the current transport route occupancy are not taken into account. It is assumed that the transporters are able to pass each other and navigate without collisions or delays. The home positions of the transport vehicles vary depending on the material supply strategy (see chapter 2.2).
Marc-André Filz et al. / Procedia CIRP 81 (2019) 1004–1009 Marc-André Filz et al. / Procedia CIRP 00 (2019) 000–000
1008
Fig. 5 gives an exemplary overview of the different positioning possibilities of the supermarket and the home positions of the transport vehicles within the MMS investigated in this paper.
5
Before modeling and evaluating different material supply strategies, target values for the evaluation of the different strategies were determined in a first series of simulation experiments without consideration of the material supply (see Table 3). During the simulation, a maximum of nine products were inside the MMS at the same time to ensure a uniform utilization. Table 3. Objective criteria for the simulation model.
Fig. 5. Positioning possibilities of supermarket.
Criteria
Objective value
Utilization manufacturing system
59.6 %
Utilization logistic system
100 %
Output
16.377
Mean value lead time P1/ P2
80.48 min / 68.95 min
For the evaluation of the material supply strategies, the lead times of the products and the utilization of the logistics system as well as the output are considered in addition to the utilization of the whole manufacturing system. Fig. 6 shows the average utilization of the manufacturing system in combination with the respective material supply strategy. 60%
3.2. Model implementation and configuration
50%
A simulation tool, which is able to combine the necessary modeling methodologies in a multi-method simulation, is the software AnyLogic®. Accordingly, this software is suitable for modeling, analysis and evaluation of the described material supply strategies in MMS. In addition, the software provides various options for the graphical visualization of model elements during a simulation run. The configuration of the model is possible regarding diverse parameters. With regard to material supply strategies within the MSS, the simulation model can be configured regarding the chosen material supply strategy, number of WS, maximum number of products to be loaded with one vehicle, number of vehicles within the system, velocity of vehicles, location of supermarket, total time of load and unload. 4. Application of simulation concept To run the simulation, all necessary parameters are read from an Excel® spreadsheet. However, the adjustment of the parameters is possible in the simulation environment. This case study presents the application of the developed simulation approach for the comparison of different material supply concepts with respect to the impact on the system utilization. The basis for the application is presented in the case study by [14]. For this simulation, the MMS consists of nine WS. Within this paper, three material supply strategies are modeled with the developed simulation tool (see Table 2). Table 2. Selected material supply strategies for simulation.
Shopping basket system Tugger train system AGV system
Routing
Allocation AGV
Position of supermarket
fix fix variable
fix fix variable
inside MMS outside MMS outside MMS
40% 30%
20% 10% 0%
Tugger train system
Shopping basket system
AGV system
Utilization of production system
Fig. 6. Simulation results of utilization of production system.
None of these material supply strategies is able to guarantee 100 % adherence to delivery dates or to deliver the required materials to the required location on time. The highest utilization result is achieved by the automated guided vehicle system with a value of 54 %. Within this simulation, the shopping basket system provides a utilization result of 50 % while the tugger train system has a value of 44 %. Table 4 gives an overview of the simulation results for output, lead time and the utilization of the respective logistic concept. The results show that the performance in terms of output and cycle times correlates with the utilization results for the specific material supply strategy. Table 4. Detailed simulation results for material supply strategies. Material supply strategy
Output
Lead time [min] P1/ P2
Utilization logistic
Tugger train system
11.316
118.81/101.06
86.7 %
Shopping basket system
13.704
97.83/84.35
50.0 %
AGV system
14.981
84.84/72.72
46.8 %
With regard to the utilization of the logistics system the tugger train system scores a result of 87 %, which is significantly higher than the AGV system with a result of 47 %
6
Marc-André Filz et al. / Procedia CIRP 81 (2019) 1004–1009 Marc-André Filz et al. / Procedia CIRP 00 (2019) 000–000
and the shopping basket system with a utilization of 50 %. If it comes to a comparison of output, the automated guided vehicle system has the highest output with 14.981 units, followed by the shopping basket system with 13.704 units and the tugger train system with 11.316 units. Comparing the lead times for product 1 and 2, the AGV system has the shortest lead time before the shopping basket system and the tugger train system. Overall, the case study has shown that the AGV system achieves the best results. Besides the highest utilization of the manufacturing system in combination with a material supply strategy, the total output of AGV system is only 8.52 % below the target value (see Table 3 and 4). However, it also becomes clear that the utilization of the AGV system is very low with a value of 46.8 %. This is due to the fact that the individual AGV are unable to level the uneven demand for transport capacity by re-allocating transport orders to AGV that are currently idle. The capacity utilization values of the logistics systems indicate that one of the causes for the failure to reach the target values is the length of the transport routes or the resulting transport time. The determinants of the supply time are the transport distance, the driving speed and the process times for loading and unloading. Moreover, the positioning of the supermarket within the manufacturing system has an impact on the utilization of the individual WS by setting the individual transport route. 5. Conclusion and outlook The developed simulation approach supports production planning by designing evaluating MMS configurations regarding different material supply strategies. Within this paper, three material supply strategies for the supply of MMS were presented and implemented within a simulation model. The results of the simulation experiments were analyzed and subsequently evaluated on the basis of selected criteria and compared with each other. None of the concepts described achieved the targeted values. The material supply via AGV proved to be the most powerful and most promising concept regarding output, lead time and utilization of the entire manufacturing system. Since the simulation results show a low utilization of the logistics system caused by length of transport routes or resulting transport time, further research is necessary to optimize the layout to shorten transport distances. Moreover, further research is needed to develop innovative and efficient concepts that meet the requirements of matrix-structured manufacturing systems. These could include intelligent routing algorithms that take into account the transport intensities within the system, further production control mechanisms as well as layout variants and their influence on the performance of material supply. References [1] Koren Y, Shpitalni M. Design of reconfigurable manufacturing systems. J Manuf Syst 2011. [2] Greschke P, Schönemann M, Thiede S, Herrmann C. Matrix structures for high volumes and flexibility in production systems. Procedia CIRP, vol. 17, 2014, p. 160–5.
1009
[3] Nyhuis P, Wiendahl H-P, Fiege T, Mühlenbruch H. Materialbereitstellung in der Montage. Montage Der Ind Produktion 2006;111:324–51. [4] Seibold Z, Furmans K. Plug&Play - Fördertechnik in der Industrie 4.0. Handb. Ind. 4.0 Band 3 - Logistik. 2., Berlin: Springer Vieweg; 2017, p. 3– 21. [5] Klug F. Steigerung der Flächenproduktivität durch Logistiksupermärkte. ZWF Zeitschrift Für Wirtschaftlichen Fabrikbetr 2012;107:72–6. [6] Klug F. Logistikmanagement in der Automobilindustrie: Grundlagen der Logistik im Automobilbau. 2. Auflage. Berlin: Springer Vieweg; 2018. [7] Ullrich G. Fahrerlose Transportsysteme. Wiesbaden: Springer Fachmedien Wiesbaden; 2014. [8] Banks J, Carson JS, Nelson BL, Nicol D. Discrete-event system simulation. Prentice Hall; 2010. [9] Zhou L, Zhang L, Ren L. Modelling and simulation of logistics service selection in cloud manufacturing. Procedia CIRP 2018;72:916–21. [10] Negahban A, Smith JS. Simulation for manufacturing system design and operation: Literature review and analysis. J Manuf Syst 2014;33:241–61. [11] Borshchev A, Filippov A. From System Dynamics and Discrete Event to Practical Agent Based Modeling: Reasons, Techniques, Tools n.d. [12] Leitão P, Barbosa J, Trentesaux D. Bio-inspired multi-agent systems for reconfigurable manufacturing systems. Eng Appl Artif Intell 2012;25:934– 44. [13] Monostori L, Váncza J, Kumara SRT. Agent-based systems for manufacturing. CIRP Ann - Manuf Technol 2006;55:697–720. [14] Schönemann M, Herrmann C, Greschke P, Thiede S. Simulation of matrix-structured manufacturing systems. J Manuf Syst 2015;37:104–12. [15] Büth L, Broderius N, Herrmann C, Thiede S. Introducing Agent-Based Simulation of Manufacturing System to Industrial Discrete-Event Simulation Tools. 2017 IEEE 15th Int Conf Ind Informatics 2017:3–8. [16] Kern W, Lämmermann H, Bauernhansl T. An Integrated Logistics Concept for a Modular Assembly System. Procedia Manuf 2017;11:957– 64. [17] Scholz M, Serno M, Franke J, Schuderer P. A Hybrid Transport Concept for the Material Supply of a Modular Manufacturing Environment. Procedia Manuf 2017;11:1448–53.
ScienceDirect ScienceDirect Available online atonline www.sciencedirect.com Procedia CIRP 00 (2019) 000–000 Available at www.sciencedirect.com
ScienceDirect ScienceDirect
Procedia CIRP 00 (2019) 000–000
www.elsevier.com/locate/procedia www.elsevier.com/locate/procedia
ProcediaProcedia CIRP 00CIRP (2017) 81000–000 (2019) 1004–1009
52nd CIRP Conference on Manufacturing Systems 52nd CIRP Conference on Manufacturing Systems
www.elsevier.com/locate/procedia
Analyzing different material supply strategies in matrix-structured Analyzing different supplyMay strategies in matrix-structured 28th CIRPmaterial Design Conference, 2018, Nantes, France manufacturing systems manufacturing systems a a A newMarc-André methodology analyze the functional and physical architecture of Filza,*,to Johann Gerberding , Christoph Herrmann , Sebastian Thiedea a,* a a Marc-André FilzTools , Johann Gerberding , Christoph Herrmannfamily , Life Sebastian Thiedea existing products for anProduction assembly product Institute of Machine and Technology, oriented Chair of Sustainable Manufacturing and Cycleidentification Engineering, a
Technische Universität Braunschweig, Langer Kamp 19 b, 38106 Braunschweig, Germany Institute of Machine Tools and Production Technology, Chair of Sustainable Manufacturing and Life Cycle Engineering, * Corresponding author. Tel+49-531-390-65032; fax: +49-531-391-5842. Technische Universität Braunschweig,E-mail Langeraddress: Kamp [email protected] b, 38106 Braunschweig, Germany a
Paul Stief *, Jean-Yves Dantan, Alain Etienne, Ali Siadat
* Corresponding author. Tel+49-531-390-65032; fax: +49-531-391-5842. E-mail address: [email protected] École Nationale Supérieure d’Arts et Métiers, Arts et Métiers ParisTech, LCFC EA 4495, 4 Rue Augustin Fresnel, Metz 57078, France
Abstract
* Corresponding author. Tel.: +33 3 87 37 54 30; E-mail address: [email protected]
Abstract The matrix-structured manufacturing system represents an alternative manufacturing concept that strives for both a volume and variant flexible production while still being efficient. The manufacturing system is a modular and cycle-independent production principle in which all existing The matrix-structured manufacturing system represents an alternative manufacturing concept that strives for both a volume and variant flexible workstations are linked together. By using intelligent decentralized control logic, the assignment of workpieces to workstations is enabled. Due Abstract production while still being efficient. The manufacturing system is a modular and cycle-independent production principle in which all existing to the functional principle of a matrix-structured manufacturing system, the material requirements at the individual workstations are not known workstations are linked together. By using intelligent decentralized control logic, the assignment of workpieces to workstations is enabled. Due advance. The behavior of different material supply strategies can be investigated with theishelp of an agent-based simulation. In this context, Inin business environment, the trend towards more productsystem, variety andmaterial customization unbroken. Due to thisworkstations development, of totoday’s the functional principle of a matrix-structured manufacturing the requirements at the individual arethe notneed known the objective of this paper is to model and analyze selected material supply strategies with regard to selected key figures in an agent-based agile and reconfigurable production systems emerged to cope with various products and product families. To design and optimize production in advance. The behavior of different material supply strategies can be investigated with the help of an agent-based simulation. In this context, simulation environment. systems as well to paper chooseisthe optimaland product matches, product analysis arewith needed. Indeed, most of knowninmethods aim to the objective ofasthis to model analyze selected material supplymethods strategies regard to selected keythefigures an agent-based analyze a product or one product family on the physical level. Different product families, however, may differ largely in terms of the number and simulation environment. © 2019 Authors. Published Elsevier This iscomparison an open access the CC BY-NC-ND license nature ofThe components. This fact by impedes anLtd. efficient and article choice under of appropriate product family combinations for the production © 2019 The Authors. Published by Elsevier Ltd. (http://creativecommons.org/licenses/by-nc-nd/3.0/) system. AThe new methodology is proposed to analyze existing products in article view ofunder theirthe functional and physical architecture. The aim is to cluster © 2019 Authors. Published by Elsevier Ltd. This is an open access CC BY-NC-ND license This is an open access article under the scientific CC BY-NC-ND license (http://creativecommons.org/licenses/by-nc-nd/3.0/) Peer-review responsibility of the committee of the 52nd CIRPofConference on Manufacturing Systems. these productsunder in new assembly oriented product families for the optimization existing assembly lines and the creation of future reconfigurable (http://creativecommons.org/licenses/by-nc-nd/3.0/) Peer-review under responsibility of the scientific committee of the 52nd CIRP Conference on Manufacturing Systems. assembly systems. on Datum Chain, the physicalofstructure the products is analyzed. FunctionalSystems. subassemblies are identified, and Peer-review underBased responsibility of Flow the scientific committee the 52ndofCIRP Conference on Manufacturing manufacturing system; Logistic concepts; Manufacturing simulation; Intelligent manufacturing systems; Flexibility a Keywords: functionalMatrix-structured analysis is performed. Moreover, a hybrid functional and physical architecture graph (HyFPAG) is the output which depicts the similarity product manufacturing families by providing design supportManufacturing to both, production planners and product designers. Keywords:between Matrix-structured system; Logistic concepts; simulation;system Intelligent manufacturing systems; FlexibilityAn illustrative example of a nail-clipper is used to explain the proposed methodology. An industrial case study on two product families of steering columns of thyssenkrupp Presta France is then carried out to give a first industrial evaluation of the proposed approach. 1. Introduction components and thus has a great influence on the efficiency of © 2017 The Authors. Published by Elsevier B.V. theDesign wholeConference production system. network structure of the Peer-review under responsibility of the scientific committee of the 28th CIRP 1. Introduction components and thus2018. has a greatThe influence on the efficiency of
In numerous industrial sectors, assembly line production is
MMS, which consists ofsystem. an uncoupled modularstructure and redundant the whole production The network of the WS design, results in multiple demand locations forredundant the same MMS, which consists of an uncoupled modular and currently exposed toprinciple. an increasing amount of areas variants, material or results modules. As a result, thelocations destinations the the main production Companies in these are WS design, in multiple demand for theofsame decreasing product life cycles and a strongly fluctuating materials are not known in advance. The production control currently exposed to an increasing amount of variants, material or modules. As a result, the destinations of the demand [1].product This results in corresponding flexibility and system assigns theknown products to the WS atproduction short notice and decreasing life cycles and a strongly fluctuating materials are not in advance. The control 1.efficiency Introduction of the product range and characteristics manufactured and/or requirements for the production principle. The depends on the particular circumstances, which limits the demand [1]. This results in corresponding flexibility and assembled system assigns the products tocontext, the WSthe at main shortchallenge notice and in this system. In this in recently developed concept of a matrix manufacturing system ability to plan the provision and increases the requirements for efficiency requirements for the production principle. The depends on the particular circumstances, which limits the Due to the fast development in the domain of modelling and analysis is now not only to cope with single provides developed the necessary high operational flexibility and flexibility and the responsiveness. recently of a matrix ability to aplan provision and increases theproduct requirements for communication andconcept an ongoing trend manufacturing of digitizationsystem and products, limited product as range families, scalability [2]. Due to the complexity wellorasexisting dynamic and stochastic provides the necessary high operational flexibility and flexibility and responsiveness. digitalization, manufacturing manufacturing enterprises are facing also to bethis ablemanufacturing to analyze andapproach, to compare to define The matrix-structured systemimportant (MMS) but behavior theproducts requirements for scalability [2]. Due toofthe complexity as well as dynamic and stochastic challenges in today’s market environments: a continuing new product families. It can be observed that classical existing consists of modular uncoupled workstations (WS) which are production logistics increase. To overcome these challenges, The matrix-structured manufacturing system (MMS) behavior of this manufacturing approach, the requirements for tendency reduction of product development times families are regrouped in function of clients or features. connectedtowards a flexible transportation system (WS) (see Fig. 1).and An product this paper develops simulation for different material consists ofby modular uncoupled workstations which are production logistics aincrease. To model overcome these challenges, shortened product lifecycles. In addition, there is an increasing However, assembly oriented product families are hardly to find. intelligent production control system coordinates the material supply strategies within the MMS with regard to selected key connected by a flexible transportation system (see Fig. 1). An this paper develops a simulation model for different material demand of customization, being at the same time in a global On the product family level, products differ mainly in two flow of the products through the production system. An performance indicators. intelligent production control system coordinates the material supply strategies within the MMS with regard to selected key competition with competitors all over the world. This trend, main characteristics: (i) the number of components and (ii) the elementary of every is the flow of thecomponent products through theproduction productionsystem system. An performance indicators. which the development from macro micro type of components (e.g. mechanical, electrical, electronical). internalis inducing material supply, whichproduction provides the torequired elementary component of every system is the markets, results in diminished lot sizes due to augmenting Classical methodologies considering mainly single products internal material supply, which provides the required product varieties (high-volume to low-volume production) [1]. or solitary, already existing product families analyze the 2212-8271 © 2019 The Authors. Published by Elsevier Ltd. This is an open access article under the CC BY-NC-ND license To cope with this augmenting variety as well as to be able to product structure on a physical level (components level) which (http://creativecommons.org/licenses/by-nc-nd/3.0/) 2212-8271 © 2019 The Authors. Published by Elsevier Ltd. This is an open access article under the CC BY-NC-ND license an efficient definition and identify possible optimization potentials in the existing causes regarding Peer-review under responsibility of the scientific committee of the 52nd CIRP Conference on difficulties Manufacturing Systems. (http://creativecommons.org/licenses/by-nc-nd/3.0/) production system, it is important to have a precise knowledge comparison of different product families. Addressing this theInmain production principle. Companies in areas are Keywords: Assembly; Design method; Family identification numerous industrial sectors, assembly linethese production is
Peer-review under responsibility of the scientific committee of the 52nd CIRP Conference on Manufacturing Systems.
2212-8271 © 2019 The Authors. Published by Elsevier Ltd. This is an©open article Published under theby CC BY-NC-ND 2212-8271 2017access The Authors. Elsevier B.V. license (http://creativecommons.org/licenses/by-nc-nd/3.0/) Peer-review under responsibility of scientific the scientific committee theCIRP 52ndDesign CIRPConference Conference2018. on Manufacturing Systems. Peer-review under responsibility of the committee of the of 28th 10.1016/j.procir.2019.03.242
2
Marc-André Filz et al. / Procedia CIRP 81 (2019) 1004–1009 Marc-André Filz et al. / Procedia CIRP 00 (2019) 000–000
2. Material supply strategies for matrix manufacturing systems 2.1. Requirements for material supply strategies In order to be able to define the requirements for material supply in MMS, the influencing factors on material supply and its design must be determined. Since material supply is a logistics system, it is part of the superordinate production system. Characteristics of the production system thus define the framework conditions for the material supply. In addition, the logistical (e.g. usage frequency), physical (e.g. volumes) and handling (e.g. bulk goods) characteristics of the material spectrum to be provided have a large influence on the design of the material supply [3]. In order to gain a better understanding of the requirements of MMS, a comparison of a line and MMS configuration is shown in Fig. 1.
1005
Table 1. Comparison of requirements regarding line and MMS configuration. Requirements
Line configuration
MMS configuration
Delivery locations
single dependant on assembly order
multiple
Material summary Short-term capacity planning
fix
hardly possible variable/ dynamic
In addition, due to lot size 1, the amount of transport operations within the MMS increases. Due to the network structure of the system, there is an overlap of the main material flow and the production supply flow. In combination with a high number of transport operations, this increases the risk of blockages on the transport routes and makes the overall system very complex. So-called deadlocks can occur especially in close-meshed layouts and represent critical situations. For example, decentrally controlled automated guided vehicles (AGV) trigger a circular closing and completely block each other and need to be prevented [4]. 2.2. Material supply strategies for MMS
Fig. 1. Material supply strategies for MMS.
The MMS network structure of unlinked modular working stations and its multiple redundant design result in multiple delivery locations for the same material or modules. The layout of the individual WS must be designed in a way that the scope of supply of the respective WS can be easily arranged. In addition, source-sink relationships in an MMS are variable and behave dynamically depending on the respective system status and the selected production control logic. Therefore, different supply locations within the production system exist for each workpiece (WP). A summary of material is thus hardly possible due to a missing planning basis. After the material infeed into the production system, the WS on which the product is processed are unknown as they are determined at short notice and depending on the situation by the production control system. This results in a missing detailed short-term capacity planning. Table 1 gives an overview of the different requirements regarding the line and MMS configuration.
The initial basis for the development of material supply strategies for MMS is a logistical segmentation regarding the article spectrum to be provided. The design of the material staging process depends largely on the application context and usually consists of a combination of different concepts, each of which is applied to a specific segment of the article spectrum. For the supply of components in an MMS, which are characterized by small batch sizes as well as a high variety of types and variants, product-specific order supply as a set or kit is recommended. Parts and modules that can be combined as a set with respect to their volumes should be delivered to the WS pre-commissioned in order to keep inventory levels within the system as low as possible. Due to the uncertainties with regard to the production schedule of a MMS, the set and module commissioning per product occurs in a supermarket close to the production. Individual supply as direct delivery is suggested in the case of large-volume parts which cannot be supplied within a set due to their volumes or geometric properties. This ensures optimum use of space and transparency within the production environment. In this context, it is necessary to consider that a just-in-sequence delivery (JIS) is not possible due to the lack of a stable order sequence in the MMS. This paper focuses on material transport between supermarket and WS within the MMS. This results in three possible material supply strategies for MMS that are shown in Fig. 2.
Fig. 2. Selected material supply strategies for MMS.
A supply by shopping basket is a form of material supply (e.g. JIT) and is frequently used in the automotive industry. The
1006
Marc-André Filz et al. / Procedia CIRP 81 (2019) 1004–1009 Marc-André Filz et al. / Procedia CIRP 00 (2019) 000–000
parts within the shopping basket are optimally arranged ergonomically according to their respective assembly order. Depending on the capacity of the shopping basket, it can provide the required material for one or more WP of a product. This material supply is particularly suitable for small and medium-sized parts with many variants [5,6]. The internal transport of the shopping basket can be carried out within the MMS by the product specific, automated and guided vehicle system. If the capacity of a shopping basket is sufficient for the number of parts to be assembled, no additional logistical effort is required. If the capacity is insufficient, it must be changed or refilled during the production process. In order to avoid longer interruptions in the production process and long transport distances, logistics stations can be integrated within the MMS. If an attached shopping basket has been exhausted, the product AGV drives to the logistics station and the attached empty shopping basket is replaced by a new shopping basket. Afterwards the product AGV drives to the next WS. The supply of the parts by tugger train system is an alternative to the shopping basket supply, which also has no restrictions regarding the part size. After a WS has been supplied with the required materials, empty containers are also returned. The number of routes and the routing within an MMS are variable to ensure that the material supply can adapt to the respective conditions and react flexibly. In addition, the performance of a tugger train system depends on the layout design, in particular the positioning of the material supply source. In the MMS, this source represents a supermarket which can be positioned within or outside the manufacturing system. An automated guided vehicle system for material supply in MMS is a highly complex system that transports the entire material spectrum of a production program. Compared to the supply by shopping basket, the AGV has no fixed allocation to a specific product. It is therefore necessary to combine different AGV in a complete logistics system. The structure of a MMS results in a complex network of different transport routes. The AGV can operate autonomously within the network for the supply and disposal of the production. Within the AGV, the selection of the vehicle depends on the materials to be transported [7]. 2.3. Simulation of matrix-structured manufacturing systems In industry and science, simulation is a well-known methodology for representing or imitating real system over time [8]. Since production logistics systems are usually quite complex, simulation is often used for the validation, analysis and optimization of the systems [9]. In the case of manufacturing, a simulation method is used as a supporting method for several tasks. Often it is used for layout design, planning, analysis and optimization of manufacturing systems [10]. Furthermore, a simulation of production can help to analyse a system with regard to its cause-effect relationships and to make the system behavior visible to the observer [8,10]. A widely used method for production simulation is discrete event simulation [10]. Therefore, passive entities, such as people or tasks, move through a manufacturing system and trigger actions at discrete events over time. Moreover, another paradigm is the approach of agent-based simulation. This
3
approach focuses on decentralized modelling of the behavior of individual entities within a predefined environment [11]. Each element within this system acts according to its own individual logic and interacts dynamically with other elements. In addition, agent-based simulation approaches can be used to model, simulate and control manufacturing systems [11–13]. Accordingly, an appropriate simulation model for the analysis of the described logistics concepts is also needed within the scope of this paper. Regarding the applications for MMS, [2] present a methodology that enables assembly line production to combine high flexibility with profitability due to the elimination of identical cycle times while maintaining a seamless process. The focus of this paper is on a systematic assignment of several operating steps to specially equipped WS and a control system that controls the corresponding distribution and ensures the dynamic configurability of the system [2]. Moreover, [14] discuss the main principles, elements and control strategies of a MMS. In addition, the paper introduces a simulation approach for the evaluation of the MMS control strategies. However, the simulation approach is applied to a use case and measures are derived on how this approach can be used for the planning of MMS. A detailed consideration of logistics is not the primary focus of the work. [15] focus on an agent-based simulation approach for industrial grade software tools to increase the flexibility of the manufacturing system. Therefore, a general methodology for implementing agent-based logic is presented based on a MMS use-case. With more focus on logistics, [16] describe an integrated logistics concept for the future modular final assembly of automobiles, consisting of five different material supply strategies. In the course of a case study in pre-assembly at a German automobile manufacturer, a space reduction of 15 % was achieved with the help of this concept. However, there is no validation or consideration of the performance of the concept. Moreover, [17] describe a hybrid transport concept for the material supply of a modular production system using tugger trains in combination with automated guided vehicles. A validation of the presented concept is not available in this context. However, none of these approaches regarding the simulation of MMS and logistics concept anticipate operational influences, like utilization or efficiency, for logistic planning and optimization. Therefore, no conclusions about the validity or operative effects of the material supply strategy on the system can be made. 3. Simulation model for the analysis of material supply concepts in matrix-structured manufacturing systems The previous investigations regarding the efficiency of the MMS assume that the production logistics is able to provide the required material at the right time at the right place in the right quality. However, due to the fact that the MMS is a highly dynamic and hardly predictable system, detailed planning of the material supply is not possible. As a result, the requirements placed on production logistics in terms of flexibility and responsiveness are increasing significantly. Therefore, a model is presented in the following, which examines the material
4
Marc-André Filz et al. / Procedia CIRP 81 (2019) 1004–1009 Marc-André Filz et al. / Procedia CIRP 00 (2019) 000–000
supply concepts described previously for their performance in a MMS. 3.1. Development of simulation model architecture The MMS simulation model within this paper consists of nine WS on which different products are manufactured. Each of the WS is able to carry out different WP and consists of a processing workplace and a buffer space with the capacity of one product (see Fig. 1). Since the production processes in a MMS can, for example, also involve assembly activities, which are usually carried out by employees, the processing times of the WP are not constant. These can vary from employee to employee for various reasons (e.g. age, qualification, learning curve), which is why the processing times are assumed to be normally distributed within the scope of this work. The material flow of the products is carried out by the selected material supply strategy (see Fig. 2). The routing to the respective successor station is determined by the product itself. It queries the status data of all possible successor stations and calculates the corresponding transport times. In accordance with [14] the WS is selected at which the time to start processing the product is the shortest. The maximum number of products within the MMS is limited to prevent blockades of the entire system [14]. Moreover, this maximum number of products within the system depends on the system state. With the help of a simulation, the optimal maximum number of products simultaneously within the MMS was determined as nine products. Fig. 3 shows the main object classes of the developed simulation model and some of the most important class parameters, variables and functions.
Fig. 3. Simplified representation of model classes.
The basic MMS concept considered within this paper is based on [14]. In addition to the products and WS, the material provision in the form of two agents for the transport system and the transport orders is integrated within this simulation architecture. The product agents are able to navigate autonomously through the production system according to the production control settings. Moreover, the product agents
1007
trigger the material supply for their next WP themselves by running an (OrderMaterial()) function after completing the previous work package. The corresponding material requirements are autonomously assigned to each WP. The order function creates transfer orders for the required material, which contain the staging location and quantity. Furthermore, it transfers them to an order list. The main parameters of the transporter agent are the type of transportation and loadable capacity. The specific behavior of the agent is modelled by a state diagram shown in Fig. 4.
Fig. 4. Statechart of transport agent.
Within this architecture, it is assumed that the material provision system ensures 100 % availability. At the start of the simulation, the transport vehicles are all waiting in their home positions. A search function (SearchOrder()) is called every five seconds to check a list of transport requests. If one or more requests exist, the transporter selects the first request from the list and executes it. In the first step, it drives to the loading station and stays there for the corresponding loading time period. Subsequently, the transfer to the staging point occurs where a certain amount of time is spent waiting during unloading before returning to the home position and selecting the next transport order. The routing of the transport vehicles takes place automatically using the shortest route within the network to the provisioning point, whereby collisions and the current transport route occupancy are not taken into account. It is assumed that the transporters are able to pass each other and navigate without collisions or delays. The home positions of the transport vehicles vary depending on the material supply strategy (see chapter 2.2).
Marc-André Filz et al. / Procedia CIRP 81 (2019) 1004–1009 Marc-André Filz et al. / Procedia CIRP 00 (2019) 000–000
1008
Fig. 5 gives an exemplary overview of the different positioning possibilities of the supermarket and the home positions of the transport vehicles within the MMS investigated in this paper.
5
Before modeling and evaluating different material supply strategies, target values for the evaluation of the different strategies were determined in a first series of simulation experiments without consideration of the material supply (see Table 3). During the simulation, a maximum of nine products were inside the MMS at the same time to ensure a uniform utilization. Table 3. Objective criteria for the simulation model.
Fig. 5. Positioning possibilities of supermarket.
Criteria
Objective value
Utilization manufacturing system
59.6 %
Utilization logistic system
100 %
Output
16.377
Mean value lead time P1/ P2
80.48 min / 68.95 min
For the evaluation of the material supply strategies, the lead times of the products and the utilization of the logistics system as well as the output are considered in addition to the utilization of the whole manufacturing system. Fig. 6 shows the average utilization of the manufacturing system in combination with the respective material supply strategy. 60%
3.2. Model implementation and configuration
50%
A simulation tool, which is able to combine the necessary modeling methodologies in a multi-method simulation, is the software AnyLogic®. Accordingly, this software is suitable for modeling, analysis and evaluation of the described material supply strategies in MMS. In addition, the software provides various options for the graphical visualization of model elements during a simulation run. The configuration of the model is possible regarding diverse parameters. With regard to material supply strategies within the MSS, the simulation model can be configured regarding the chosen material supply strategy, number of WS, maximum number of products to be loaded with one vehicle, number of vehicles within the system, velocity of vehicles, location of supermarket, total time of load and unload. 4. Application of simulation concept To run the simulation, all necessary parameters are read from an Excel® spreadsheet. However, the adjustment of the parameters is possible in the simulation environment. This case study presents the application of the developed simulation approach for the comparison of different material supply concepts with respect to the impact on the system utilization. The basis for the application is presented in the case study by [14]. For this simulation, the MMS consists of nine WS. Within this paper, three material supply strategies are modeled with the developed simulation tool (see Table 2). Table 2. Selected material supply strategies for simulation.
Shopping basket system Tugger train system AGV system
Routing
Allocation AGV
Position of supermarket
fix fix variable
fix fix variable
inside MMS outside MMS outside MMS
40% 30%
20% 10% 0%
Tugger train system
Shopping basket system
AGV system
Utilization of production system
Fig. 6. Simulation results of utilization of production system.
None of these material supply strategies is able to guarantee 100 % adherence to delivery dates or to deliver the required materials to the required location on time. The highest utilization result is achieved by the automated guided vehicle system with a value of 54 %. Within this simulation, the shopping basket system provides a utilization result of 50 % while the tugger train system has a value of 44 %. Table 4 gives an overview of the simulation results for output, lead time and the utilization of the respective logistic concept. The results show that the performance in terms of output and cycle times correlates with the utilization results for the specific material supply strategy. Table 4. Detailed simulation results for material supply strategies. Material supply strategy
Output
Lead time [min] P1/ P2
Utilization logistic
Tugger train system
11.316
118.81/101.06
86.7 %
Shopping basket system
13.704
97.83/84.35
50.0 %
AGV system
14.981
84.84/72.72
46.8 %
With regard to the utilization of the logistics system the tugger train system scores a result of 87 %, which is significantly higher than the AGV system with a result of 47 %
6
Marc-André Filz et al. / Procedia CIRP 81 (2019) 1004–1009 Marc-André Filz et al. / Procedia CIRP 00 (2019) 000–000
and the shopping basket system with a utilization of 50 %. If it comes to a comparison of output, the automated guided vehicle system has the highest output with 14.981 units, followed by the shopping basket system with 13.704 units and the tugger train system with 11.316 units. Comparing the lead times for product 1 and 2, the AGV system has the shortest lead time before the shopping basket system and the tugger train system. Overall, the case study has shown that the AGV system achieves the best results. Besides the highest utilization of the manufacturing system in combination with a material supply strategy, the total output of AGV system is only 8.52 % below the target value (see Table 3 and 4). However, it also becomes clear that the utilization of the AGV system is very low with a value of 46.8 %. This is due to the fact that the individual AGV are unable to level the uneven demand for transport capacity by re-allocating transport orders to AGV that are currently idle. The capacity utilization values of the logistics systems indicate that one of the causes for the failure to reach the target values is the length of the transport routes or the resulting transport time. The determinants of the supply time are the transport distance, the driving speed and the process times for loading and unloading. Moreover, the positioning of the supermarket within the manufacturing system has an impact on the utilization of the individual WS by setting the individual transport route. 5. Conclusion and outlook The developed simulation approach supports production planning by designing evaluating MMS configurations regarding different material supply strategies. Within this paper, three material supply strategies for the supply of MMS were presented and implemented within a simulation model. The results of the simulation experiments were analyzed and subsequently evaluated on the basis of selected criteria and compared with each other. None of the concepts described achieved the targeted values. The material supply via AGV proved to be the most powerful and most promising concept regarding output, lead time and utilization of the entire manufacturing system. Since the simulation results show a low utilization of the logistics system caused by length of transport routes or resulting transport time, further research is necessary to optimize the layout to shorten transport distances. Moreover, further research is needed to develop innovative and efficient concepts that meet the requirements of matrix-structured manufacturing systems. These could include intelligent routing algorithms that take into account the transport intensities within the system, further production control mechanisms as well as layout variants and their influence on the performance of material supply. References [1] Koren Y, Shpitalni M. Design of reconfigurable manufacturing systems. J Manuf Syst 2011. [2] Greschke P, Schönemann M, Thiede S, Herrmann C. Matrix structures for high volumes and flexibility in production systems. Procedia CIRP, vol. 17, 2014, p. 160–5.
1009
[3] Nyhuis P, Wiendahl H-P, Fiege T, Mühlenbruch H. Materialbereitstellung in der Montage. Montage Der Ind Produktion 2006;111:324–51. [4] Seibold Z, Furmans K. Plug&Play - Fördertechnik in der Industrie 4.0. Handb. Ind. 4.0 Band 3 - Logistik. 2., Berlin: Springer Vieweg; 2017, p. 3– 21. [5] Klug F. Steigerung der Flächenproduktivität durch Logistiksupermärkte. ZWF Zeitschrift Für Wirtschaftlichen Fabrikbetr 2012;107:72–6. [6] Klug F. Logistikmanagement in der Automobilindustrie: Grundlagen der Logistik im Automobilbau. 2. Auflage. Berlin: Springer Vieweg; 2018. [7] Ullrich G. Fahrerlose Transportsysteme. Wiesbaden: Springer Fachmedien Wiesbaden; 2014. [8] Banks J, Carson JS, Nelson BL, Nicol D. Discrete-event system simulation. Prentice Hall; 2010. [9] Zhou L, Zhang L, Ren L. Modelling and simulation of logistics service selection in cloud manufacturing. Procedia CIRP 2018;72:916–21. [10] Negahban A, Smith JS. Simulation for manufacturing system design and operation: Literature review and analysis. J Manuf Syst 2014;33:241–61. [11] Borshchev A, Filippov A. From System Dynamics and Discrete Event to Practical Agent Based Modeling: Reasons, Techniques, Tools n.d. [12] Leitão P, Barbosa J, Trentesaux D. Bio-inspired multi-agent systems for reconfigurable manufacturing systems. Eng Appl Artif Intell 2012;25:934– 44. [13] Monostori L, Váncza J, Kumara SRT. Agent-based systems for manufacturing. CIRP Ann - Manuf Technol 2006;55:697–720. [14] Schönemann M, Herrmann C, Greschke P, Thiede S. Simulation of matrix-structured manufacturing systems. J Manuf Syst 2015;37:104–12. [15] Büth L, Broderius N, Herrmann C, Thiede S. Introducing Agent-Based Simulation of Manufacturing System to Industrial Discrete-Event Simulation Tools. 2017 IEEE 15th Int Conf Ind Informatics 2017:3–8. [16] Kern W, Lämmermann H, Bauernhansl T. An Integrated Logistics Concept for a Modular Assembly System. Procedia Manuf 2017;11:957– 64. [17] Scholz M, Serno M, Franke J, Schuderer P. A Hybrid Transport Concept for the Material Supply of a Modular Manufacturing Environment. Procedia Manuf 2017;11:1448–53.