- Email: [email protected]
Computer Simulation and Optimization of Transport Distances of Order Picking Processes
Available online at www.sciencedirect.com
ScienceDirect Procedia Engineering 192 (2017) 69 – 74
TRANSCOM 2017: International scientific conference on sustainable, modern and safe transport
Computer simulation and optimization of transport distances of order picking processes 0RQLND%XþNRYia*0DUWLQ.UDMþRYLþa, Milan Edlb a
University of Žilina, Faculty of Mechanical Engineering, Department of Industrial Engineering, University 8215/1, 010 26 Žilina, Slovakia University of West Bohemia, Faculty of Mechanical Engineering, Department of Industrial Engineering and Management, Univerzitní ul., þ. orientaþní 8, þ.p. 2732, 306 14 PlzeĖ, Czech republic
b
Abstract This article deals with optimizing of transport distances in orders picking processes. Article shows solution how to optimize orders picking in warehouse using dynamic simulation. In article basic steps of order picking planning process are described in order to minimize distances traveled by truck or human. The basis of this solution is execution of computer simulation and simulation experiments for faster finding of correct solution, which causes company cost reduction. 2017 2017The TheAuthors. Authors. Published Elsevier © Published by by Elsevier Ltd.Ltd. This is an open access article under the CC BY-NC-ND license Peer-review under responsibility of the scientific committee of TRANSCOM 2017: International scientific conference on (http://creativecommons.org/licenses/by-nc-nd/4.0/). Peer-review responsibility of the scientific committee of TRANSCOM 2017: International scientific conference on sustainable, sustainable,under modern and safe transport. modern and safe transport Keywords: Picking processes ; Simulation ; Experiments ; Transport
1. Optimization of order picking processes in warehouses with using dynamic simulation Storage and its correct functioning has significant impact on ensuring higher level of customer service and in protecting utility properties of goods. The method and speed of goods picking is one of the most important areas in warehouses. Rapid development of technologies and market growth causes that companies must deal with questions of satisfying difficult customers with specific and individual requirements and brings more questions about creation of inventories, which complicates designing of warehouses, data collection, picking of orders according to making
* Corresponding author. Tel.: + 421-(0) 41 513 2701; fax: + 421-(0) 41 513 1501. E-mail address: [email protected]
1877-7058 © 2017 The Authors. Published by Elsevier Ltd. This is an open access article under the CC BY-NC-ND license
(http://creativecommons.org/licenses/by-nc-nd/4.0/). Peer-review under responsibility of the scientific committee of TRANSCOM 2017: International scientific conference on sustainable, modern and safe transport
doi:10.1016/j.proeng.2017.06.012
70
Monika Bučková et al. / Procedia Engineering 192 (2017) 69 – 74
changes and improving logistics in warehouses [1]. Today factories cant wait to decide if changes that they realize will bring necessary results. Each waiting is a loss and every loss entails costs that every manager must eliminate in manufacturing and warehouse [2]. One of many tools that allows managers to make decisions is computer simulation and solutions of digital factory, which has wide application range in companies. In this article it will described possibility of using computer simulation to optimize transport distances of order picking processes in warehouse by worker on practical example created in software Tecnomatix Plant Simulation 13. 1.1. Simulation model of order picking processes Monitoring and planning processes of orders picking in warehouses is important step, because computer simulation can point out when the system begins to fail, when it will be overwhelmed by goods, how to find optimal route for workers during picking orders. This is the way how to reduce number of steps that worker does, thereby shorten time of picking operations. In this case, it is necessary to design experiments and monitor how system will react on changes, for example changing number of workers, reordering sequence of orders or changes in routes. It is also necessary to determine utilization of workers in warehouses, what the core activity of their work is and how much time they spend on the fulfillment different activities. Orders picking can be performed in individual orders or in batches of orders. Single batch may contain any number of individual orders, which for some reason is useful to process as a one-time batch. One of the reason may be requirement for picking several orders at same time - for a particular carrier or for a network of carriers. According to these reasons, it is advantageous to use computer simulation as a tool to optimize orders picking operations. Computer simulation works with simulation model created on computer, it will be time consuming and very costly to capture and understand all processes that take place in the company or warehouse. Before creation of the model itself, it is necessary to determine only relevant processes and relationships that effect on given system in aspect of main objective, which must be achieved through simulation experiments. Phase of experiments is usually the most interesting phase of the project, because it begins to bring first results. Statistical analysis is an integral part of each simulation project. It can create debate just about results of experimentation moreover bringing new ideas and possibilities for improving simulation model.
Fig. 1. Model of order picking processes by worker.
In displayed case in Fig.1, we can consider to find solution of order picking by worker during one working shift. Worker's task is to prepare order, composed from items called V1, V2, V3, V4, V5, which are placed in racks with special places numbered S16, S4, S8, S15, S1. Main goal of simulation is described so we can make experiments with this model.
71
Monika Bučková et al. / Procedia Engineering 192 (2017) 69 – 74
Fig. 2. 3D view of simulation model of order picking processes by worker.
For this example (Fig. 1., Fig. 2.) we have chosen lower number of orders due to large number of possibilities of combinations. To solve this problem we can use the basics of permutations without repetition, in M is a set of n different elements, which constitute n - tuple, with elements in n - tuples can not be repeated. ܲሺ݊ሻ ൌ ݊Ǩ
(1)
By five orders called V1, V2, V3, V4, V5. Each order is placed on one of five locations S16, S4, S8, S15, S1, we can consider with 120 possibilities of possible routes. We defined all the routes so in experiments we can find the shortest possible route to help worker to find this items as quickly as possible. Table 1. Table of combinations storage locations S16, S4, S8, S15, S1. Generating permutations of a set of storage locations S16, S4, S8, S15, S1 1
4
8
15 16 4
1
16 15 4
4
1
8
15 16 15 4
8
1
4
15 16 15 1
8
1
8
4
1
8
4
15 16 1
15 8
4
16 15 1
4
8
1
15 16 8
15 1
4
16 16 1
8 8
4
1
15 16 15 8
4
4
15 1
16 1
8
15 4
4
8
15 1
16 8
1
15 4
16 16 4
1
4
1
16 4
1
15 8
16 1
4
16 8 15 4
8
8
15 4
1
16 1
4
15 8
16 4
1
16 8 15 8
4
1
15 16 4
8
4
15 1
16 1 8
16 15 8
15 8
15 8
1
1
8 16 4
1
8
16 15 16 4
16 4
15 1
8 16 1
4
8
16 1
15 4
8 16 8
4
1
4
8 16 4
8
1
16 15 16 1
8
4
8 15 1
8
4
16 15 8
1
16 4
15 16 15 8
4
1
16 1
16 4
8 15 8
1
4
16 15 1
8
16 4
15 15 16 8
4
1
16 4
16 1
8 15 8
16 4
1
4
15 8
16 15 4
1
16 1
4
15 16 8
15 4
1
16 1
15 15 16 1
4
8
16 4
15 8
1
16 1
15 16 15 1
4
8
4
16 15 8
1
4
15 8
1
16 4
15 16 4
8
1
8
1
15 16 8
1
16 8
15 4
8 15 16 8
1
4
8
1
15 8
16 15 4
16 8
1
15 15 8
4
16 15 1
16 8
4
15 15 8
16 4
1
4
15 8
15 16 4
1
1
15 16 8
1
15 8
15 4
16 1 16 1
15 1
16 4
8
15 4
1
16 8
4
15 16 1 8
16 1
15 8
4
8
15 1
16 4
16 15 4
16 1
15 16 1 4
15 1
15 8
16 4
4
16 4
15 4
16 8
15 4
8
1
4
15 16 8
8
16 8
4
1
1
16 15 4
8
4
1
15 16 8
15 8
16 1 4
1
15 16 8
4
15 1
8
1
16 4
4
16 8
16 8
15 1 4
1
8
16 15 4
8
1
15 16 4
8
4
16 15 1
15 4
16 8
8
16 15 1 4
8
1
16 15 4
1
8
15 16 4
4
8
16 15 1
15 8
15 1
15 8
1
1
16 1
4
4
1
16 15 8
16 15 4
1
8
15 16 8
1 4
16 1
8
15 4
4
8
15 16 1
16 8
4
15 1
1
4
16 15 8
15 16 4
1
8
16 15 8
1 4
1
16 8
15 4
8
4
15 16 1
8
16 4
15 1
1
15 8
4
16 15 1
8
16 15 1
8 4
8
16 1
15 4
15 4
8
16 1
4
16 8
15 1
16 1
15 8
16 4
8
15 16 1
8 4
16 8
15 4
4
15 8
16 1
16 4
16 4 4
15 1
1
8
15 1
72
Monika Bučková et al. / Procedia Engineering 192 (2017) 69 – 74
All the combinations shown in Table 1. must be evaluate. For faster evaluation of such a large number of experiments, we can use tool from software Tecnomatix Plant Simulation 13 - Experiment Manager, because every event that is occurred during simulation, is automatically recorded, which simplifies work and evaluation of results. Larger amount of created variants of dynamic system causes higher probability to find the most suitable solution. 1.2. Experimenting with model by using simulation tool Tecnomatix Plant Simulation 13
Fig. 3. Sample of the output report created with Experiments Manager (a, b).
After entering input set of combinations and their values into the tool - Experiment Manager we can define output variable and software automatically perform experiments and results are displayed in the form of summary tables and graphs. The optimization can be performed in next steps using genetic algorithms. Tecnomatix Plant Simulation 13 software displays models in 2D mode, 3D mode, or 2D / 3D mode. The latest version of this software allows you to present concept of system created in full virtual, more interactive environment and work with higher amount of tools for statistical evaluation of simulation models. Using Experiment Manager is possible to create output of simulation model in form of web document (Fig. 3. a). Advantage of this displaying is that it is possible send results of simulation immediately using various web solutions e.g. by e-mail. The most important part of this output is automatically create graph - shown on Fig. 4.
Fig. 4. A graph of comparison Experiment ± Distance.
Monika Bučková et al. / Procedia Engineering 192 (2017) 69 – 74
On this chart, Fig. 4. are on the x-axis showing experiments and on the y-axis are captured distances what worker had to go while picking orders with items V1, V2, V3, V4, V5. Graph, Fig. 4. (a), is automatically created by the described software, on the Fig. 4. (b), are marked by red lines experiments Exp044 and Exp119. When we compare graph on the Fig. 4 (b). and Table 1, on Fig.1, we can say that the worker has passed the shortest distance displayed on experiment Exp044 on way between S16 - S1 - S8 - S4 - S15 and the longest in experiment Exp119 due to combination S4 - S16 - S8 - S15 - S1. In this way it is possible to evaluate distances, which are not passed only by workers, but also by various handling devices and their combinations.
Fig. 5. Detailed 3D view of (a) worker and (b) automatic recording and monitoring distance what worker passed.
It is possible to animate objects (Fig. 5.) and their individual elements created in 3D models in Tecnomatix Plant Simulation 13 software. This step is important because warehouse operations must be highly detailed reprocess. Tecnomatix Plant Simulation module for warehousing and logistics is tool suitable for redesigning of existing processes. Reliable models could be developed relatively quickly and alternative scenarios and another proposals can be created easier [4]. This example is created as sample for lower number of orders. It is difficult to recalculate all permutations without repetition and immediately determine route for worker or for handling equipment when number of location will increase. Therefore, one of possible way to solve these optimization problems is to use genetic algorithms. 1.3. Genetic algorithms The genetic algorithm we could described as non-deterministic method of problem solving, based on the principle of Darwin's theory of evolution. We could include them to stochastic optimization techniques. Problem solution, in this case, is called a chromosome and is constituted by binary string of a given length, which is the same for all chromosomes of the population. The final set of chromosomes is called as a population. Opening balance of problem solving is called basic population. Development of optimal solutions to solve the problem is caused by natural population development. Randomness is ensured by generating a pseudorandom numbers. Each created population by reproduction is improved by the rating of chromosomes with using the function f (x) called fitness. In the process of solving the problem using genetic algorithms, it comes to finding local maximum of fitness function, thus we need to find best rated solution to solve the problem in state space [5]. The algorithms used for generating solving optimization tasks is an iterative process. Simulation model evaluates suitability of proposed solutions. Connection of simulation model created in Tecnomatix Plant Simulation 13 and genetic algorithms it is possible to use an iterative process to find new and better solutions. After optimization cycles, which software allows to make, the model is still ready for further use. The main advantage of using genetic algorithms
73
74
Monika Bučková et al. / Procedia Engineering 192 (2017) 69 – 74
to solve optimization problems of picking operations, whether workers or handling equipment is that, they are not dependent on tasks [5]. In this practical sample system is firmly set. In the case, if model will contain random components, which entered into it, for example equipment fails, it is necessary to make several number of simulation runs for each newly created subject. 2. Conclusion In created practical sample of solution how to optimize transport routes during orders picking using computer simulation, we can see how tedious process of recalculation could be. Results of computer simulation can show us distances that worker had to go by picking defined order. This problem with small number of orders can be solved by calculating permutations without repetition and using Experiment Manager from simulation software Tecnomatix Plant Simulation 13. The higher amount of orders makes it more time consuming and tediously for user to determine the shortest and longest transport route for picking orders. Therefore, one of the suitable tools for optimization of transport distances should be genetic algorithms connected to simulation models. Its structure and basics supports further development of simulation models. To create and using genetic algorithms with support of simulation software is necessary to own professional version of the software. What is the major disadvantage for smaller companies, because genetic algorithms modules are mostly the most expensive additional libraries of simulation software. Simulation of orders picking can be important tool also in terms of ergonomics, i.e. removal of unnecessary steps and shortening wasted time, when worker has to find items, to bend or to stretch for it, allows worker be better focus on work. Improving ergonomics of work, thus for example shortening transport routes, leading to increased productivity, reducing the risk of occupational accidents, work-related diseases, which can lead to example employing another worker on substitute, resulting to cost increasing [6]. By using Tecnomatix Plant Simulation 13 software that is used on Department of Industrial Engineering we can create interactive 3D model of warehouse, optimize picking and transport routes, while getting very compendious results that can be easily stored into web documents. This method of data processing is fast and secure, it is a good presentation tool which can be used anywhere. Acknowledgements This article was created with support RISURMHFW.(*$ä8-4/2016. References >@-+QiW-+HUþNR0*UHJRU0RGHUQDSSURDFKWRWKHGHVLJQand control of logistic processes, in: XIII. international logistics & supply chain FRQJUHVV ³0DULWLPHORJLVWLFV WKHQHZSRUWSURMHFWVRI7XUNH\ SURFHHGLQJV -23 October 2015, Izmir, Turkey, Izmir: Izmir University & Logistics Association (LODER), 2015, ISBN 978-605-84194-2-1, pp. 229-237. >@$âWHIiQLN3*U]QiU%0LþLHWD7RROVIRU&RQWLQXDO3URFHVV,PSURYHPHQW± Simulation and Benchmarking, in: Annals of DAAM for 2003 & Proc. of the 14th Intern. DAAAM Symposium: Intelligent manufacturing & automation: Focus on reconstruction and development, 2003, pp. 443-444, ISBN 978-3-901509-34-6. >@%0LþLHWDď'XOLQD00DOFKR0DLQIDFWRUVRIWKHVHOHFWLRQMREVIRUWKHZRUNVWXG\LQ $QQDOVRI'$$$0IRU 3roceedings of the 16th International DAAAM Symposium: Manufacturing & automation: Focus on young researches and scientists, 2005, pp. 249-250, ISBN: 978-3-901509-46-9. >@6'LOVNê1iYUKV\VWpPXLQWHUDNWtYQHKRORJLVWLFNpKRSOiQRYDQLDäLOLQVNiXQLYHU]LWDYäLOLQH6WURMQtFNDIDNXOWD.DWHGUDSULHP\VHOQpKR LQåLQLHUVWYDý923ULHP\VHOQpLQåLQLHUVWYRäLOLQD >VQ@ S&'520 >@0*UHJRU-+HUþNR3*U]QiU7KHIDFWRU\RIWKHIXWXUHSURGXFWLRQV\VWHPUHVHDUFKLQ ,&$&3URFHHGLQJVRIWKH1-st International conference on automation and computing, Glasgow, UK, September 11 ± 12, 2015, [S.l.]: IEEE, 2015, ISBN 978-0-9926801-0-7. >@%0LþLHWDď'XOLQD00DOFKR2SWLPDOXWLOL]DWLRQRIWKHHPSOR\HHVSRWHQWLRQLQWKHPDQXIDFWXULQJFRQGLWLRQLQ $QQDls of DAAAM for 2003 & Proceedings of the 14th International DAAAM Symposium: Intelligent manufacturing & automation: Focus on reconstruction and development, 2003, pp.129 ± 130, ISBN: 978-3-901509-34-6.
ScienceDirect Procedia Engineering 192 (2017) 69 – 74
TRANSCOM 2017: International scientific conference on sustainable, modern and safe transport
Computer simulation and optimization of transport distances of order picking processes 0RQLND%XþNRYia*0DUWLQ.UDMþRYLþa, Milan Edlb a
University of Žilina, Faculty of Mechanical Engineering, Department of Industrial Engineering, University 8215/1, 010 26 Žilina, Slovakia University of West Bohemia, Faculty of Mechanical Engineering, Department of Industrial Engineering and Management, Univerzitní ul., þ. orientaþní 8, þ.p. 2732, 306 14 PlzeĖ, Czech republic
b
Abstract This article deals with optimizing of transport distances in orders picking processes. Article shows solution how to optimize orders picking in warehouse using dynamic simulation. In article basic steps of order picking planning process are described in order to minimize distances traveled by truck or human. The basis of this solution is execution of computer simulation and simulation experiments for faster finding of correct solution, which causes company cost reduction. 2017 2017The TheAuthors. Authors. Published Elsevier © Published by by Elsevier Ltd.Ltd. This is an open access article under the CC BY-NC-ND license Peer-review under responsibility of the scientific committee of TRANSCOM 2017: International scientific conference on (http://creativecommons.org/licenses/by-nc-nd/4.0/). Peer-review responsibility of the scientific committee of TRANSCOM 2017: International scientific conference on sustainable, sustainable,under modern and safe transport. modern and safe transport Keywords: Picking processes ; Simulation ; Experiments ; Transport
1. Optimization of order picking processes in warehouses with using dynamic simulation Storage and its correct functioning has significant impact on ensuring higher level of customer service and in protecting utility properties of goods. The method and speed of goods picking is one of the most important areas in warehouses. Rapid development of technologies and market growth causes that companies must deal with questions of satisfying difficult customers with specific and individual requirements and brings more questions about creation of inventories, which complicates designing of warehouses, data collection, picking of orders according to making
* Corresponding author. Tel.: + 421-(0) 41 513 2701; fax: + 421-(0) 41 513 1501. E-mail address: [email protected]
1877-7058 © 2017 The Authors. Published by Elsevier Ltd. This is an open access article under the CC BY-NC-ND license
(http://creativecommons.org/licenses/by-nc-nd/4.0/). Peer-review under responsibility of the scientific committee of TRANSCOM 2017: International scientific conference on sustainable, modern and safe transport
doi:10.1016/j.proeng.2017.06.012
70
Monika Bučková et al. / Procedia Engineering 192 (2017) 69 – 74
changes and improving logistics in warehouses [1]. Today factories cant wait to decide if changes that they realize will bring necessary results. Each waiting is a loss and every loss entails costs that every manager must eliminate in manufacturing and warehouse [2]. One of many tools that allows managers to make decisions is computer simulation and solutions of digital factory, which has wide application range in companies. In this article it will described possibility of using computer simulation to optimize transport distances of order picking processes in warehouse by worker on practical example created in software Tecnomatix Plant Simulation 13. 1.1. Simulation model of order picking processes Monitoring and planning processes of orders picking in warehouses is important step, because computer simulation can point out when the system begins to fail, when it will be overwhelmed by goods, how to find optimal route for workers during picking orders. This is the way how to reduce number of steps that worker does, thereby shorten time of picking operations. In this case, it is necessary to design experiments and monitor how system will react on changes, for example changing number of workers, reordering sequence of orders or changes in routes. It is also necessary to determine utilization of workers in warehouses, what the core activity of their work is and how much time they spend on the fulfillment different activities. Orders picking can be performed in individual orders or in batches of orders. Single batch may contain any number of individual orders, which for some reason is useful to process as a one-time batch. One of the reason may be requirement for picking several orders at same time - for a particular carrier or for a network of carriers. According to these reasons, it is advantageous to use computer simulation as a tool to optimize orders picking operations. Computer simulation works with simulation model created on computer, it will be time consuming and very costly to capture and understand all processes that take place in the company or warehouse. Before creation of the model itself, it is necessary to determine only relevant processes and relationships that effect on given system in aspect of main objective, which must be achieved through simulation experiments. Phase of experiments is usually the most interesting phase of the project, because it begins to bring first results. Statistical analysis is an integral part of each simulation project. It can create debate just about results of experimentation moreover bringing new ideas and possibilities for improving simulation model.
Fig. 1. Model of order picking processes by worker.
In displayed case in Fig.1, we can consider to find solution of order picking by worker during one working shift. Worker's task is to prepare order, composed from items called V1, V2, V3, V4, V5, which are placed in racks with special places numbered S16, S4, S8, S15, S1. Main goal of simulation is described so we can make experiments with this model.
71
Monika Bučková et al. / Procedia Engineering 192 (2017) 69 – 74
Fig. 2. 3D view of simulation model of order picking processes by worker.
For this example (Fig. 1., Fig. 2.) we have chosen lower number of orders due to large number of possibilities of combinations. To solve this problem we can use the basics of permutations without repetition, in M is a set of n different elements, which constitute n - tuple, with elements in n - tuples can not be repeated. ܲሺ݊ሻ ൌ ݊Ǩ
(1)
By five orders called V1, V2, V3, V4, V5. Each order is placed on one of five locations S16, S4, S8, S15, S1, we can consider with 120 possibilities of possible routes. We defined all the routes so in experiments we can find the shortest possible route to help worker to find this items as quickly as possible. Table 1. Table of combinations storage locations S16, S4, S8, S15, S1. Generating permutations of a set of storage locations S16, S4, S8, S15, S1 1
4
8
15 16 4
1
16 15 4
4
1
8
15 16 15 4
8
1
4
15 16 15 1
8
1
8
4
1
8
4
15 16 1
15 8
4
16 15 1
4
8
1
15 16 8
15 1
4
16 16 1
8 8
4
1
15 16 15 8
4
4
15 1
16 1
8
15 4
4
8
15 1
16 8
1
15 4
16 16 4
1
4
1
16 4
1
15 8
16 1
4
16 8 15 4
8
8
15 4
1
16 1
4
15 8
16 4
1
16 8 15 8
4
1
15 16 4
8
4
15 1
16 1 8
16 15 8
15 8
15 8
1
1
8 16 4
1
8
16 15 16 4
16 4
15 1
8 16 1
4
8
16 1
15 4
8 16 8
4
1
4
8 16 4
8
1
16 15 16 1
8
4
8 15 1
8
4
16 15 8
1
16 4
15 16 15 8
4
1
16 1
16 4
8 15 8
1
4
16 15 1
8
16 4
15 15 16 8
4
1
16 4
16 1
8 15 8
16 4
1
4
15 8
16 15 4
1
16 1
4
15 16 8
15 4
1
16 1
15 15 16 1
4
8
16 4
15 8
1
16 1
15 16 15 1
4
8
4
16 15 8
1
4
15 8
1
16 4
15 16 4
8
1
8
1
15 16 8
1
16 8
15 4
8 15 16 8
1
4
8
1
15 8
16 15 4
16 8
1
15 15 8
4
16 15 1
16 8
4
15 15 8
16 4
1
4
15 8
15 16 4
1
1
15 16 8
1
15 8
15 4
16 1 16 1
15 1
16 4
8
15 4
1
16 8
4
15 16 1 8
16 1
15 8
4
8
15 1
16 4
16 15 4
16 1
15 16 1 4
15 1
15 8
16 4
4
16 4
15 4
16 8
15 4
8
1
4
15 16 8
8
16 8
4
1
1
16 15 4
8
4
1
15 16 8
15 8
16 1 4
1
15 16 8
4
15 1
8
1
16 4
4
16 8
16 8
15 1 4
1
8
16 15 4
8
1
15 16 4
8
4
16 15 1
15 4
16 8
8
16 15 1 4
8
1
16 15 4
1
8
15 16 4
4
8
16 15 1
15 8
15 1
15 8
1
1
16 1
4
4
1
16 15 8
16 15 4
1
8
15 16 8
1 4
16 1
8
15 4
4
8
15 16 1
16 8
4
15 1
1
4
16 15 8
15 16 4
1
8
16 15 8
1 4
1
16 8
15 4
8
4
15 16 1
8
16 4
15 1
1
15 8
4
16 15 1
8
16 15 1
8 4
8
16 1
15 4
15 4
8
16 1
4
16 8
15 1
16 1
15 8
16 4
8
15 16 1
8 4
16 8
15 4
4
15 8
16 1
16 4
16 4 4
15 1
1
8
15 1
72
Monika Bučková et al. / Procedia Engineering 192 (2017) 69 – 74
All the combinations shown in Table 1. must be evaluate. For faster evaluation of such a large number of experiments, we can use tool from software Tecnomatix Plant Simulation 13 - Experiment Manager, because every event that is occurred during simulation, is automatically recorded, which simplifies work and evaluation of results. Larger amount of created variants of dynamic system causes higher probability to find the most suitable solution. 1.2. Experimenting with model by using simulation tool Tecnomatix Plant Simulation 13
Fig. 3. Sample of the output report created with Experiments Manager (a, b).
After entering input set of combinations and their values into the tool - Experiment Manager we can define output variable and software automatically perform experiments and results are displayed in the form of summary tables and graphs. The optimization can be performed in next steps using genetic algorithms. Tecnomatix Plant Simulation 13 software displays models in 2D mode, 3D mode, or 2D / 3D mode. The latest version of this software allows you to present concept of system created in full virtual, more interactive environment and work with higher amount of tools for statistical evaluation of simulation models. Using Experiment Manager is possible to create output of simulation model in form of web document (Fig. 3. a). Advantage of this displaying is that it is possible send results of simulation immediately using various web solutions e.g. by e-mail. The most important part of this output is automatically create graph - shown on Fig. 4.
Fig. 4. A graph of comparison Experiment ± Distance.
Monika Bučková et al. / Procedia Engineering 192 (2017) 69 – 74
On this chart, Fig. 4. are on the x-axis showing experiments and on the y-axis are captured distances what worker had to go while picking orders with items V1, V2, V3, V4, V5. Graph, Fig. 4. (a), is automatically created by the described software, on the Fig. 4. (b), are marked by red lines experiments Exp044 and Exp119. When we compare graph on the Fig. 4 (b). and Table 1, on Fig.1, we can say that the worker has passed the shortest distance displayed on experiment Exp044 on way between S16 - S1 - S8 - S4 - S15 and the longest in experiment Exp119 due to combination S4 - S16 - S8 - S15 - S1. In this way it is possible to evaluate distances, which are not passed only by workers, but also by various handling devices and their combinations.
Fig. 5. Detailed 3D view of (a) worker and (b) automatic recording and monitoring distance what worker passed.
It is possible to animate objects (Fig. 5.) and their individual elements created in 3D models in Tecnomatix Plant Simulation 13 software. This step is important because warehouse operations must be highly detailed reprocess. Tecnomatix Plant Simulation module for warehousing and logistics is tool suitable for redesigning of existing processes. Reliable models could be developed relatively quickly and alternative scenarios and another proposals can be created easier [4]. This example is created as sample for lower number of orders. It is difficult to recalculate all permutations without repetition and immediately determine route for worker or for handling equipment when number of location will increase. Therefore, one of possible way to solve these optimization problems is to use genetic algorithms. 1.3. Genetic algorithms The genetic algorithm we could described as non-deterministic method of problem solving, based on the principle of Darwin's theory of evolution. We could include them to stochastic optimization techniques. Problem solution, in this case, is called a chromosome and is constituted by binary string of a given length, which is the same for all chromosomes of the population. The final set of chromosomes is called as a population. Opening balance of problem solving is called basic population. Development of optimal solutions to solve the problem is caused by natural population development. Randomness is ensured by generating a pseudorandom numbers. Each created population by reproduction is improved by the rating of chromosomes with using the function f (x) called fitness. In the process of solving the problem using genetic algorithms, it comes to finding local maximum of fitness function, thus we need to find best rated solution to solve the problem in state space [5]. The algorithms used for generating solving optimization tasks is an iterative process. Simulation model evaluates suitability of proposed solutions. Connection of simulation model created in Tecnomatix Plant Simulation 13 and genetic algorithms it is possible to use an iterative process to find new and better solutions. After optimization cycles, which software allows to make, the model is still ready for further use. The main advantage of using genetic algorithms
73
74
Monika Bučková et al. / Procedia Engineering 192 (2017) 69 – 74
to solve optimization problems of picking operations, whether workers or handling equipment is that, they are not dependent on tasks [5]. In this practical sample system is firmly set. In the case, if model will contain random components, which entered into it, for example equipment fails, it is necessary to make several number of simulation runs for each newly created subject. 2. Conclusion In created practical sample of solution how to optimize transport routes during orders picking using computer simulation, we can see how tedious process of recalculation could be. Results of computer simulation can show us distances that worker had to go by picking defined order. This problem with small number of orders can be solved by calculating permutations without repetition and using Experiment Manager from simulation software Tecnomatix Plant Simulation 13. The higher amount of orders makes it more time consuming and tediously for user to determine the shortest and longest transport route for picking orders. Therefore, one of the suitable tools for optimization of transport distances should be genetic algorithms connected to simulation models. Its structure and basics supports further development of simulation models. To create and using genetic algorithms with support of simulation software is necessary to own professional version of the software. What is the major disadvantage for smaller companies, because genetic algorithms modules are mostly the most expensive additional libraries of simulation software. Simulation of orders picking can be important tool also in terms of ergonomics, i.e. removal of unnecessary steps and shortening wasted time, when worker has to find items, to bend or to stretch for it, allows worker be better focus on work. Improving ergonomics of work, thus for example shortening transport routes, leading to increased productivity, reducing the risk of occupational accidents, work-related diseases, which can lead to example employing another worker on substitute, resulting to cost increasing [6]. By using Tecnomatix Plant Simulation 13 software that is used on Department of Industrial Engineering we can create interactive 3D model of warehouse, optimize picking and transport routes, while getting very compendious results that can be easily stored into web documents. This method of data processing is fast and secure, it is a good presentation tool which can be used anywhere. Acknowledgements This article was created with support RISURMHFW.(*$ä8-4/2016. References >@-+QiW-+HUþNR0*UHJRU0RGHUQDSSURDFKWRWKHGHVLJQand control of logistic processes, in: XIII. international logistics & supply chain FRQJUHVV ³0DULWLPHORJLVWLFV WKHQHZSRUWSURMHFWVRI7XUNH\ SURFHHGLQJV -23 October 2015, Izmir, Turkey, Izmir: Izmir University & Logistics Association (LODER), 2015, ISBN 978-605-84194-2-1, pp. 229-237. >@$âWHIiQLN3*U]QiU%0LþLHWD7RROVIRU&RQWLQXDO3URFHVV,PSURYHPHQW± Simulation and Benchmarking, in: Annals of DAAM for 2003 & Proc. of the 14th Intern. DAAAM Symposium: Intelligent manufacturing & automation: Focus on reconstruction and development, 2003, pp. 443-444, ISBN 978-3-901509-34-6. >@%0LþLHWDď'XOLQD00DOFKR0DLQIDFWRUVRIWKHVHOHFWLRQMREVIRUWKHZRUNVWXG\LQ $QQDOVRI'$$$0IRU 3roceedings of the 16th International DAAAM Symposium: Manufacturing & automation: Focus on young researches and scientists, 2005, pp. 249-250, ISBN: 978-3-901509-46-9. >@6'LOVNê1iYUKV\VWpPXLQWHUDNWtYQHKRORJLVWLFNpKRSOiQRYDQLDäLOLQVNiXQLYHU]LWDYäLOLQH6WURMQtFNDIDNXOWD.DWHGUDSULHP\VHOQpKR LQåLQLHUVWYDý923ULHP\VHOQpLQåLQLHUVWYRäLOLQD >VQ@ S&'520 >@0*UHJRU-+HUþNR3*U]QiU7KHIDFWRU\RIWKHIXWXUHSURGXFWLRQV\VWHPUHVHDUFKLQ ,&$&3URFHHGLQJVRIWKH1-st International conference on automation and computing, Glasgow, UK, September 11 ± 12, 2015, [S.l.]: IEEE, 2015, ISBN 978-0-9926801-0-7. >@%0LþLHWDď'XOLQD00DOFKR2SWLPDOXWLOL]DWLRQRIWKHHPSOR\HHVSRWHQWLRQLQWKHPDQXIDFWXULQJFRQGLWLRQLQ $QQDls of DAAAM for 2003 & Proceedings of the 14th International DAAAM Symposium: Intelligent manufacturing & automation: Focus on reconstruction and development, 2003, pp.129 ± 130, ISBN: 978-3-901509-34-6.