- Email: [email protected]
Computer Simulation as a Tool for the Optimization of Logistics Using Automated Guided Vehicles
Available online at www.sciencedirect.com
ScienceDirect Procedia Engineering 192 (2017) 923 – 928
TRANSCOM 2017: International scientific conference on sustainable, modern and safe transport
Computer simulation as a tool for the optimization of logistics using automated guided vehicles Vladimír Vavríka*, Milan Gregora, Patrik Grznára a
University of Žilina, Faculty of Mechanical Engineering, Department of Industrial Engineering, Univerzitná 1, 010 26 Žilina, Slovak Republic
Abstract The article describes results of the research project and at the same time, it introduces the method of the determination of number of automated guided vehicles and choosing of optimal internal company logistics track. New technologies are fundamentally changing the internal logistics and internal logistics is therefore gradually becoming adaptive, and that requires changes in the whole concept of future solutions. One example is automated logistics system of planned operation of manufacturing semi-products intra-process of components production in the automotive industry. The simulation results of the logistics system were variants for increasing the use of the operation areas, optimized material supply and created layout that would be able to flexibly response to the future company requirements. 2017The The Authors. Published by Elsevier Ltd.is an open access article under the CC BY-NC-ND license ©©2017 Authors. Published by Elsevier Ltd. This Peer-review under responsibility ofthe scientific committee of TRANSCOM 2017: International scientific conference on (http://creativecommons.org/licenses/by-nc-nd/4.0/). Peer-review responsibility the scientific committee of TRANSCOM 2017: International scientific conference on sustainable, sustainable,under modern and safe of transport. modern and safe transport Keywords: Computer simulation; automated guided vehicle; automated logistics system; plant Simulation
1. Introduction The utilization of computer simulation considerably supports production planning and control. It is one of the main parts of the digital factory. The simulation enables the imitation of a suggested solution to determine the system’s parameters in order to reach requested goals. One of the primary goals of each company is to increase the effectivity of the particular processes by using simulation. Simulation enables imitate process in production area, logistics,
* Corresponding author. Tel.: +421-41-513-2713; fax: +421-041-513-1501 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.2017.06.159
924
Vladimír Vavrík et al. / Procedia Engineering 192 (2017) 923 – 928
assembly etc. The article deals with the internal logistics in the factory, which aims into production volume growth. Such changes will require a new internal logistics routes. Tecnomatix Plant Simulation software from SIEMENS Corporation was used in this simulation project of new logistics. The Software is a component of PLM, whereas it allows analysis of all process in discrete time. [1,2] 2. Targets definition and problem analysis of the factory The logistics situation in the factory is a combination of automated logistics and typical forklift transport. Growing market demand resulting in the product volume growth caused the extension of one of the production halls and purchase of new machines for semi–finished products processing. This process influenced localization of former technological process in the factory and induced need for the new logistics track. To achieve effective transport by using new track it is necessary to choose a suitable logistics vehicle and to optimize transport track. For logistics, the transport with automated guided vehicles was chosen, which substantially reduces needs of the labor force and provide effective transport of the semi-finished product. After the type of transport has been chosen, it was necessary to design the track itself, so three different variants were proposed. These proposed tracks had to be evaluated. Simulation and bottlenecks analysis were used to evaluate the actual production conditions. However, the main aim of all simulation experiments was to estimate a required number of transport vehicles, which must be implemented into the operation of the business. [3,4] Simulation project in this article follows the methods, where the concept of simulation model needs to be preceded by static calculation of necessary number of vehicles. This calculation is a base for dynamic simulation and it can be used only for initial validation of the basic model. The more exact are input parameters, the more exact are the results of the dynamic simulation, and so the results of simulation experiments usually differ from static calculation (Fig. 1).
Fig. 1. The method of simulation model development for the factory (Source: own construction).
Vladimír Vavrík et al. / Procedia Engineering 192 (2017) 923 – 928
It has to be mentioned that a conceptual model proposal was based on a collection of data from the working environment where it was necessary to assess value of current state and value of future production state. Future state is important to be taken into account considering that the growth of production output of the company in following years is expected. 3. Static calculation of vehicles number and data collection As it was mentioned above it is suitable to use the static calculation of the number of vehicles using the mathematical formulas introduced in Figure 1. According to the given mathematical formula, one can assess the number of required vehicles. However, the calculation has to be done before the data collection. The company was operated in a three-shift operation mode, where expected disposable working time of vehicle per shift is 7,5 hour. Vehicles will recharge at fast charging station during worker´s half hour break. Another data that need to be calculated are speed parameters of vehicles and loading and unloading time. The approximate duration of loading/unloading process time according to corporate was 25 second, whereas the vehicle speed was set on 1,2 m/s on a straight track and on 0,8 m/s on a curve. The speed of vehicles was based on already introduced automated system in the company. Transport of two different semi–finish products was required, while 23 pieces of A semi–finish products and B semi– finished products were required to be transported within one work shift which is together 108 circuits for one vehicle per day on condition that that one vehicle is able to transport only one semi–finished product. As by the following year increased production rate is expected, the number of circuits by one vehicle per day has to be increased as well, whereas the quantity of circuits may grow up to 66 circuits per day. 4. Simulation model design The software solution itself has to be preceded by proposal of conceptual model with all important data. At this point a mathematical–logical model was proposed which will serve as a direct fundament for simulation model realization. The design of conceptual models has to be abstract without real entities as they are not important for model making process. The structure and behavior of proposed entities have to imitate those of real objects and of the whole system. The realized logical model can be subsequently reflected into selected simulation software. Creation of simulation model is subordinated to the method which is closely connected to the chosen software solution. Software solution from Tecnomatix (Fig. 2) requires the sequence of following steps for simulation model creation [5]:
Fig. 2. 2D and 3D visualization simulation model in Plant Simulation (Source: own construction).
925
926
Vladimír Vavrík et al. / Procedia Engineering 192 (2017) 923 – 928
4.1. Creation of tracks for vehicle move In contrast to static calculation, dynamic simulation takes into account random states that can occur in the real work process as well as technology of transport facility and existing layout. The logistics track in the company will be created bi–directionally because of the workplace shortage in organization layout of company. So it is necessary to prevent such collision in particular by establishing stabling tracks in selected logistics route. The problematic section can be also loaded, as the vehicle has to go back in the loop to the same direction in which they have arrived to loading. The main reason for such mechanism is, again, workplace shortage in company layout. 4.2. Creation of loading and unloading station The main step for simulation is the creation of an input entity. This entity in our model is represented by imitation of semi–finished products from real system. The vehicles are expected to ensure the transport of two semi–finished products with different parameters, which need to be transported in the specialized pallet. There arises a need to create two different pallet types (it is necessary to distinguish empty and full transport pallet, with a reason statistic in used simulation software) for two different types of semi–finished product. Regarding to the real manufacturing layout, the model has to contain two loading/unloading stations of full specialized pallets and two loading/unloading stations of empty specialized pallet (one loading station of full specialized pallets can be used only for loading of semi-finished product A), which is together 8 transfers stations. 4.3. Creation of vehicle for transport of semi-finished product Vehicle entity has to copy the behavior of future vehicle in the logistics system as much as possible. It is then necessary to adjust vehicle’s size parameters and also to define its maximum speed. Vehicle speed limit for automated system changes according to properties of the track on which the vehicle moves. This is also necessary to take into account in making the simulation, so the maximum speed on straight and curve needs to be changed in all variants of track. The specific value is mentioned by static calculation of a number of the vehicles. 4.4. Creation of logical rules for model The most important part of simulation model building is a creation of logical rules which represent required behavior of vehicles in the real manufacturing. It is necessary to prevent the vehicles from a collision in track during simulation, to define time and place to stop on the fast charging station and also to direct the flow of material according to the requirements of manufacturing. 4.5. Verification and validation of created model Verification means logical check of the model, carried out by observing of individual variants of the model and of entity function itself. Verification process enables both fine–tune potential deviation in existing models and avoiding incorrect results. The achieved verified results are followed by validation, the i.e. comparison of the results of real and of simulated manufacturing systems. Particular model is not realized for the status quo system, so that is why data were used from static calculation and the results from simulated model were compared with this data. 5. Simulation experiments The result of simulation of proposed logistics tracks should be simultaneously evaluated and compared with the proposed variant tracks. This has to result into determination of the optimum number of the vehicle for every variant. However, it is necessary to take into account the status quo and the expected status of manufacturing in the future because of logistics transport is supposed to increase up to 66 pieces. A set of experiments was carried out depending on the change of model factors.
Vladimír Vavrík et al. / Procedia Engineering 192 (2017) 923 – 928
927
x Experiments defining necessary number of vehicles x Experiments evaluating proposed tracks x Experiments simulating changes in system Adequate simulation results can be achieved only on the condition of optimal input factors of simulation model, as they say, garbage in - garbage out at simulation. The designed models were focused only on transport of materials by means of vehicles, so it was assumed that the models would not be influenced by such factors as storage capacity, machine failures or the production intensity of a manufacturing facility. Simulation is therefore limited only by logistics process. The input factor can be chosen as the constant value, which exceeds the constant value of output. If consumption of semi-finished products on output is 36 pieces/ per work shift, as an entry constant of value higher than the previous number was chosen. The material input was not generated into the system in time intervals, but rather it was gathered in buffers and in a specific ratio it was allocated on individual loading depending on requirements. Vehicles enter the system only after the entry of entities of semi-finished products because system begins to work from the state of "start from zero". The number of entering vehicles was set in initial experiments on 1 vehicle. The number of vehicles was increased and monitored for each variant. The quantity of transported material in full and empty pallets (in which the material is transported) was increased and monitored. The final data were summarized in the chart for particular variants tracks and number of AGV. The monitored parameter was intersection of transported number of semi–finished products (pallets) for the specific number of vehicles in the specific variant. [6,7] One can establish the number of vehicles in the specific variant from the resulting values which were be able to transport necessary amount of material. Next experiments were focused on the observation of individual track variants, where the change of vehicle’s performance and waiting time of vehicles was monitored resulting from loading/unloading or waiting time which does not add any value (for example: waiting time in stabling tracks, waiting time at collision with another vehicle in the crossroad, waiting time at collision with other system parts). Individual values were taken from results of many of the simulation runs and these new data were subsequently compared among themselves. Apart from the runs with original system parameters, the simulation enables also to create parametric model that will have the ability of flexible response on changes. In designing conceptual (still non-existing) system, it has to be taken into account the modification of some parameters of systemic parts. These parameters are mainly data concerning the loading/unloading time of material and vehicle speed in realized simulation model. They can change depending on used type of vehicle and loading/unloading technology of semi–finish products. There was monitored the deviation in necessary number of vehicles and original system parameters during experiments, as well as the decrease or increase of vehicle performance for given variants and number of vehicles. The change of loading and unloading time was realized by increasing or decreasing of original value per 10 seconds. After original value change (25 seconds) the growth of the quantity of transported products was observed and from this data it was possible to find out the number of AGV sufficient for ensuring transport. The results of simulation runs show also the vehicle’s performance and their waiting time where the values are averaged on 1 vehicle. Like this, we can get clearer results and determine advantages and disadvantages of individual variants for a specific number of vehicles. The vehicle speed is parameter depending on used vehicle type. Their specific parameters play important role in final assessment of number of vehicles but also in choosing the optimal logistics tracks. The company does not need to be decided for specific vehicle type before the simulation project, therefore it is necessary to choose from the options on market and on the basis of technical specification to adjust the attributes of simulation model. In the model, the original speed of vehicle was set on value 1,2 m/ per second on straight and 0,8 m/ per second in the curve. Two adjustments of speed parameters were carried out as follows: x The vehicle’s speed was decreased on 1 m/ per second on straight and 0,6 meter/ per second on the curve x The vehicle’s speed was increased on 2 m/ per second on straight and 1,6 meter/ per second on the curve Similarly as in previous experiments, it was necessary to observe the quantity of transported production for specific number of vehicles and to define the necessary number for every variant. Then the performance parameters and waiting time of vehicles were checked. The data from experiments were subsequently visualised in the chart which was used as the tool for evaluation of suitability of proposed tracks and to determination of necessary number of vehicles for the tracks (Fig. 3). In this way it was possible to create the optimal logistics track in manufacturing and
928
Vladimír Vavrík et al. / Procedia Engineering 192 (2017) 923 – 928
to eliminate increased costs by the system implementation. It is also possible to adjust the simulation models depending on the final solution of implemented system and to use simulation model as an optimization tool. Development count of vehicles on specific system settings for future state of production (p)
Development count of vehicles on specific system settings for actual state of production (p) 4
Count of vehicles
Count of vehicles
5
3 2 1 0 The basic model settings
Modification of the Modification of the Modification of the Modification of the vehicle speed on 2 vehicle speed on 1 time for time for loading/unloading on loading/unloading on m/sec. - 1,6 m/sec. m/sec. - 0,6 m/sec. 15 sec. 35 sec.
Variant of A track
Variant of B track
Variant of C track
7 6 5 4 3 2 1 0 The basic model settings
Modification of the Modification of the Modification of the Modification of the vehicle speed on 2 vehicle speed on 1 time for time for loading/unloading loading/unloading m/sec. - 1,6 m/sec. m/sec. - 0,6 m/sec. on 15 sec. on 35 sec. Variant of A track
Variant of B track
Variant of C track
Fig. 3. Graphical representation required count of vehicles from the simulation experiments and basic model (Source: own construction).
4. Conclusion It was necessary to take into account the achieved results from all carried experiments for the recommendation of required number of vehicles and choosing suitable variants. It was also necessary to individually evaluate each variant with its advantages and disadvantages versus others variants and to determine suitable number of vehicles for these variants. Besides the results of simulation runs it was necessary to take into account that the system was created in existing production, so it was required to use the above mentioned criteria connected with the real situation in production by evaluation. Before the decision concerning the choice of implementation of any variant, the company had to reassess the profitability of invested capital to given project. The simulation model of future logistics system enables factory to do fast verification of financial demand project automatization, to suggest necessary number of AVGs and allows optimization of vehicles tracks. Because of unpredictable occurrences, the simulation results can be different from results in real manufacturing, therefore it is convenient to fill simulation model with the new data during implementation and so particularize results from experiments. Acknowledgements This paper was supported by research project „Reconfigurable Logistics System for Manufacturing Systems of the New Generation of Factory of The Future (RLS_FoF)“, No. APVV-14-0752, co-financed by the Slovak Agency for R&D Support. References [1] M. Gregor, a kol. 2006. Digitálny podnik. 1. Vyd. Žilina :Slovenské centrum produktivity, 2006. pp. 148. ISBN 80-969391-5-7. (in Slovak) [2] M. Krajčovič, G. Gabajová, B. Mičieta, 2014. Order picking using augmented reality. In: Communications – Scientific letters of the University of Žilina. Vol. 16, no. 3A (2014), pp. 106-111. ISSN 1335-4205. [3] D. Plinta, M. Krajčovič, 2016. Production system designing with the use of digital factory and augmented reality technologies. In: Advances in Intelligent Systems and Computing., Vol. 350 (2016), pp. 187-196. ISSN 2194-5357. [4] M. Krajčovič, et al. 2013. Intelligent manufacturing systems in concept of digital factory In: Communications – Scientific letters of the University of Žilina. Vol. 15, no. 2 (2013), pp. 77-87. ISSN 1335-4205. [5] M. Gregor, J. Košturiak, M. Halušková, 1997. Priemyselné inžinierstvo: Simulácia výrobných systémov. Žilina : Jozef BLAHA, 1997. pp. 166 .ISBN 80-966996-8-7. [6] B. Mičieta, Ľ. Dulina, M. Malcho, 2005. Main factors of the selection jobs for the work study. In: Annals of DAAAM for 2005 & Proceedings of the 16th International DAAAM Symposium: Manufacturing & automation: Focus on young researches and scientists, 2005. pp. 249-250. ISBN: 978-3-901509-46-9. [7] B. Mičieta, M. Gašo, M. Krajčovič, 2014. Innovation performance of organization. In: Communications – Scientific letters of the University of Žilina. Vol. 16, no. 3A (2014), pp. 112-118. ISSN 1335-4205. [8] S. Bangsow, 2010. Manufacturing Simulation with plant simulation and simtalk. Berlin Heidelberg : Springer, 2010. pp. 297. ISBN 978-3642-05073-2. [9] Siemens. 2010. Tecnomatix Plant Simulation 10 Step-by-Step Help. Program manual. pp. 578.
ScienceDirect Procedia Engineering 192 (2017) 923 – 928
TRANSCOM 2017: International scientific conference on sustainable, modern and safe transport
Computer simulation as a tool for the optimization of logistics using automated guided vehicles Vladimír Vavríka*, Milan Gregora, Patrik Grznára a
University of Žilina, Faculty of Mechanical Engineering, Department of Industrial Engineering, Univerzitná 1, 010 26 Žilina, Slovak Republic
Abstract The article describes results of the research project and at the same time, it introduces the method of the determination of number of automated guided vehicles and choosing of optimal internal company logistics track. New technologies are fundamentally changing the internal logistics and internal logistics is therefore gradually becoming adaptive, and that requires changes in the whole concept of future solutions. One example is automated logistics system of planned operation of manufacturing semi-products intra-process of components production in the automotive industry. The simulation results of the logistics system were variants for increasing the use of the operation areas, optimized material supply and created layout that would be able to flexibly response to the future company requirements. 2017The The Authors. Published by Elsevier Ltd.is an open access article under the CC BY-NC-ND license ©©2017 Authors. Published by Elsevier Ltd. This Peer-review under responsibility ofthe scientific committee of TRANSCOM 2017: International scientific conference on (http://creativecommons.org/licenses/by-nc-nd/4.0/). Peer-review responsibility the scientific committee of TRANSCOM 2017: International scientific conference on sustainable, sustainable,under modern and safe of transport. modern and safe transport Keywords: Computer simulation; automated guided vehicle; automated logistics system; plant Simulation
1. Introduction The utilization of computer simulation considerably supports production planning and control. It is one of the main parts of the digital factory. The simulation enables the imitation of a suggested solution to determine the system’s parameters in order to reach requested goals. One of the primary goals of each company is to increase the effectivity of the particular processes by using simulation. Simulation enables imitate process in production area, logistics,
* Corresponding author. Tel.: +421-41-513-2713; fax: +421-041-513-1501 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.2017.06.159
924
Vladimír Vavrík et al. / Procedia Engineering 192 (2017) 923 – 928
assembly etc. The article deals with the internal logistics in the factory, which aims into production volume growth. Such changes will require a new internal logistics routes. Tecnomatix Plant Simulation software from SIEMENS Corporation was used in this simulation project of new logistics. The Software is a component of PLM, whereas it allows analysis of all process in discrete time. [1,2] 2. Targets definition and problem analysis of the factory The logistics situation in the factory is a combination of automated logistics and typical forklift transport. Growing market demand resulting in the product volume growth caused the extension of one of the production halls and purchase of new machines for semi–finished products processing. This process influenced localization of former technological process in the factory and induced need for the new logistics track. To achieve effective transport by using new track it is necessary to choose a suitable logistics vehicle and to optimize transport track. For logistics, the transport with automated guided vehicles was chosen, which substantially reduces needs of the labor force and provide effective transport of the semi-finished product. After the type of transport has been chosen, it was necessary to design the track itself, so three different variants were proposed. These proposed tracks had to be evaluated. Simulation and bottlenecks analysis were used to evaluate the actual production conditions. However, the main aim of all simulation experiments was to estimate a required number of transport vehicles, which must be implemented into the operation of the business. [3,4] Simulation project in this article follows the methods, where the concept of simulation model needs to be preceded by static calculation of necessary number of vehicles. This calculation is a base for dynamic simulation and it can be used only for initial validation of the basic model. The more exact are input parameters, the more exact are the results of the dynamic simulation, and so the results of simulation experiments usually differ from static calculation (Fig. 1).
Fig. 1. The method of simulation model development for the factory (Source: own construction).
Vladimír Vavrík et al. / Procedia Engineering 192 (2017) 923 – 928
It has to be mentioned that a conceptual model proposal was based on a collection of data from the working environment where it was necessary to assess value of current state and value of future production state. Future state is important to be taken into account considering that the growth of production output of the company in following years is expected. 3. Static calculation of vehicles number and data collection As it was mentioned above it is suitable to use the static calculation of the number of vehicles using the mathematical formulas introduced in Figure 1. According to the given mathematical formula, one can assess the number of required vehicles. However, the calculation has to be done before the data collection. The company was operated in a three-shift operation mode, where expected disposable working time of vehicle per shift is 7,5 hour. Vehicles will recharge at fast charging station during worker´s half hour break. Another data that need to be calculated are speed parameters of vehicles and loading and unloading time. The approximate duration of loading/unloading process time according to corporate was 25 second, whereas the vehicle speed was set on 1,2 m/s on a straight track and on 0,8 m/s on a curve. The speed of vehicles was based on already introduced automated system in the company. Transport of two different semi–finish products was required, while 23 pieces of A semi–finish products and B semi– finished products were required to be transported within one work shift which is together 108 circuits for one vehicle per day on condition that that one vehicle is able to transport only one semi–finished product. As by the following year increased production rate is expected, the number of circuits by one vehicle per day has to be increased as well, whereas the quantity of circuits may grow up to 66 circuits per day. 4. Simulation model design The software solution itself has to be preceded by proposal of conceptual model with all important data. At this point a mathematical–logical model was proposed which will serve as a direct fundament for simulation model realization. The design of conceptual models has to be abstract without real entities as they are not important for model making process. The structure and behavior of proposed entities have to imitate those of real objects and of the whole system. The realized logical model can be subsequently reflected into selected simulation software. Creation of simulation model is subordinated to the method which is closely connected to the chosen software solution. Software solution from Tecnomatix (Fig. 2) requires the sequence of following steps for simulation model creation [5]:
Fig. 2. 2D and 3D visualization simulation model in Plant Simulation (Source: own construction).
925
926
Vladimír Vavrík et al. / Procedia Engineering 192 (2017) 923 – 928
4.1. Creation of tracks for vehicle move In contrast to static calculation, dynamic simulation takes into account random states that can occur in the real work process as well as technology of transport facility and existing layout. The logistics track in the company will be created bi–directionally because of the workplace shortage in organization layout of company. So it is necessary to prevent such collision in particular by establishing stabling tracks in selected logistics route. The problematic section can be also loaded, as the vehicle has to go back in the loop to the same direction in which they have arrived to loading. The main reason for such mechanism is, again, workplace shortage in company layout. 4.2. Creation of loading and unloading station The main step for simulation is the creation of an input entity. This entity in our model is represented by imitation of semi–finished products from real system. The vehicles are expected to ensure the transport of two semi–finished products with different parameters, which need to be transported in the specialized pallet. There arises a need to create two different pallet types (it is necessary to distinguish empty and full transport pallet, with a reason statistic in used simulation software) for two different types of semi–finished product. Regarding to the real manufacturing layout, the model has to contain two loading/unloading stations of full specialized pallets and two loading/unloading stations of empty specialized pallet (one loading station of full specialized pallets can be used only for loading of semi-finished product A), which is together 8 transfers stations. 4.3. Creation of vehicle for transport of semi-finished product Vehicle entity has to copy the behavior of future vehicle in the logistics system as much as possible. It is then necessary to adjust vehicle’s size parameters and also to define its maximum speed. Vehicle speed limit for automated system changes according to properties of the track on which the vehicle moves. This is also necessary to take into account in making the simulation, so the maximum speed on straight and curve needs to be changed in all variants of track. The specific value is mentioned by static calculation of a number of the vehicles. 4.4. Creation of logical rules for model The most important part of simulation model building is a creation of logical rules which represent required behavior of vehicles in the real manufacturing. It is necessary to prevent the vehicles from a collision in track during simulation, to define time and place to stop on the fast charging station and also to direct the flow of material according to the requirements of manufacturing. 4.5. Verification and validation of created model Verification means logical check of the model, carried out by observing of individual variants of the model and of entity function itself. Verification process enables both fine–tune potential deviation in existing models and avoiding incorrect results. The achieved verified results are followed by validation, the i.e. comparison of the results of real and of simulated manufacturing systems. Particular model is not realized for the status quo system, so that is why data were used from static calculation and the results from simulated model were compared with this data. 5. Simulation experiments The result of simulation of proposed logistics tracks should be simultaneously evaluated and compared with the proposed variant tracks. This has to result into determination of the optimum number of the vehicle for every variant. However, it is necessary to take into account the status quo and the expected status of manufacturing in the future because of logistics transport is supposed to increase up to 66 pieces. A set of experiments was carried out depending on the change of model factors.
Vladimír Vavrík et al. / Procedia Engineering 192 (2017) 923 – 928
927
x Experiments defining necessary number of vehicles x Experiments evaluating proposed tracks x Experiments simulating changes in system Adequate simulation results can be achieved only on the condition of optimal input factors of simulation model, as they say, garbage in - garbage out at simulation. The designed models were focused only on transport of materials by means of vehicles, so it was assumed that the models would not be influenced by such factors as storage capacity, machine failures or the production intensity of a manufacturing facility. Simulation is therefore limited only by logistics process. The input factor can be chosen as the constant value, which exceeds the constant value of output. If consumption of semi-finished products on output is 36 pieces/ per work shift, as an entry constant of value higher than the previous number was chosen. The material input was not generated into the system in time intervals, but rather it was gathered in buffers and in a specific ratio it was allocated on individual loading depending on requirements. Vehicles enter the system only after the entry of entities of semi-finished products because system begins to work from the state of "start from zero". The number of entering vehicles was set in initial experiments on 1 vehicle. The number of vehicles was increased and monitored for each variant. The quantity of transported material in full and empty pallets (in which the material is transported) was increased and monitored. The final data were summarized in the chart for particular variants tracks and number of AGV. The monitored parameter was intersection of transported number of semi–finished products (pallets) for the specific number of vehicles in the specific variant. [6,7] One can establish the number of vehicles in the specific variant from the resulting values which were be able to transport necessary amount of material. Next experiments were focused on the observation of individual track variants, where the change of vehicle’s performance and waiting time of vehicles was monitored resulting from loading/unloading or waiting time which does not add any value (for example: waiting time in stabling tracks, waiting time at collision with another vehicle in the crossroad, waiting time at collision with other system parts). Individual values were taken from results of many of the simulation runs and these new data were subsequently compared among themselves. Apart from the runs with original system parameters, the simulation enables also to create parametric model that will have the ability of flexible response on changes. In designing conceptual (still non-existing) system, it has to be taken into account the modification of some parameters of systemic parts. These parameters are mainly data concerning the loading/unloading time of material and vehicle speed in realized simulation model. They can change depending on used type of vehicle and loading/unloading technology of semi–finish products. There was monitored the deviation in necessary number of vehicles and original system parameters during experiments, as well as the decrease or increase of vehicle performance for given variants and number of vehicles. The change of loading and unloading time was realized by increasing or decreasing of original value per 10 seconds. After original value change (25 seconds) the growth of the quantity of transported products was observed and from this data it was possible to find out the number of AGV sufficient for ensuring transport. The results of simulation runs show also the vehicle’s performance and their waiting time where the values are averaged on 1 vehicle. Like this, we can get clearer results and determine advantages and disadvantages of individual variants for a specific number of vehicles. The vehicle speed is parameter depending on used vehicle type. Their specific parameters play important role in final assessment of number of vehicles but also in choosing the optimal logistics tracks. The company does not need to be decided for specific vehicle type before the simulation project, therefore it is necessary to choose from the options on market and on the basis of technical specification to adjust the attributes of simulation model. In the model, the original speed of vehicle was set on value 1,2 m/ per second on straight and 0,8 m/ per second in the curve. Two adjustments of speed parameters were carried out as follows: x The vehicle’s speed was decreased on 1 m/ per second on straight and 0,6 meter/ per second on the curve x The vehicle’s speed was increased on 2 m/ per second on straight and 1,6 meter/ per second on the curve Similarly as in previous experiments, it was necessary to observe the quantity of transported production for specific number of vehicles and to define the necessary number for every variant. Then the performance parameters and waiting time of vehicles were checked. The data from experiments were subsequently visualised in the chart which was used as the tool for evaluation of suitability of proposed tracks and to determination of necessary number of vehicles for the tracks (Fig. 3). In this way it was possible to create the optimal logistics track in manufacturing and
928
Vladimír Vavrík et al. / Procedia Engineering 192 (2017) 923 – 928
to eliminate increased costs by the system implementation. It is also possible to adjust the simulation models depending on the final solution of implemented system and to use simulation model as an optimization tool. Development count of vehicles on specific system settings for future state of production (p)
Development count of vehicles on specific system settings for actual state of production (p) 4
Count of vehicles
Count of vehicles
5
3 2 1 0 The basic model settings
Modification of the Modification of the Modification of the Modification of the vehicle speed on 2 vehicle speed on 1 time for time for loading/unloading on loading/unloading on m/sec. - 1,6 m/sec. m/sec. - 0,6 m/sec. 15 sec. 35 sec.
Variant of A track
Variant of B track
Variant of C track
7 6 5 4 3 2 1 0 The basic model settings
Modification of the Modification of the Modification of the Modification of the vehicle speed on 2 vehicle speed on 1 time for time for loading/unloading loading/unloading m/sec. - 1,6 m/sec. m/sec. - 0,6 m/sec. on 15 sec. on 35 sec. Variant of A track
Variant of B track
Variant of C track
Fig. 3. Graphical representation required count of vehicles from the simulation experiments and basic model (Source: own construction).
4. Conclusion It was necessary to take into account the achieved results from all carried experiments for the recommendation of required number of vehicles and choosing suitable variants. It was also necessary to individually evaluate each variant with its advantages and disadvantages versus others variants and to determine suitable number of vehicles for these variants. Besides the results of simulation runs it was necessary to take into account that the system was created in existing production, so it was required to use the above mentioned criteria connected with the real situation in production by evaluation. Before the decision concerning the choice of implementation of any variant, the company had to reassess the profitability of invested capital to given project. The simulation model of future logistics system enables factory to do fast verification of financial demand project automatization, to suggest necessary number of AVGs and allows optimization of vehicles tracks. Because of unpredictable occurrences, the simulation results can be different from results in real manufacturing, therefore it is convenient to fill simulation model with the new data during implementation and so particularize results from experiments. Acknowledgements This paper was supported by research project „Reconfigurable Logistics System for Manufacturing Systems of the New Generation of Factory of The Future (RLS_FoF)“, No. APVV-14-0752, co-financed by the Slovak Agency for R&D Support. References [1] M. Gregor, a kol. 2006. Digitálny podnik. 1. Vyd. Žilina :Slovenské centrum produktivity, 2006. pp. 148. ISBN 80-969391-5-7. (in Slovak) [2] M. Krajčovič, G. Gabajová, B. Mičieta, 2014. Order picking using augmented reality. In: Communications – Scientific letters of the University of Žilina. Vol. 16, no. 3A (2014), pp. 106-111. ISSN 1335-4205. [3] D. Plinta, M. Krajčovič, 2016. Production system designing with the use of digital factory and augmented reality technologies. In: Advances in Intelligent Systems and Computing., Vol. 350 (2016), pp. 187-196. ISSN 2194-5357. [4] M. Krajčovič, et al. 2013. Intelligent manufacturing systems in concept of digital factory In: Communications – Scientific letters of the University of Žilina. Vol. 15, no. 2 (2013), pp. 77-87. ISSN 1335-4205. [5] M. Gregor, J. Košturiak, M. Halušková, 1997. Priemyselné inžinierstvo: Simulácia výrobných systémov. Žilina : Jozef BLAHA, 1997. pp. 166 .ISBN 80-966996-8-7. [6] B. Mičieta, Ľ. Dulina, M. Malcho, 2005. Main factors of the selection jobs for the work study. In: Annals of DAAAM for 2005 & Proceedings of the 16th International DAAAM Symposium: Manufacturing & automation: Focus on young researches and scientists, 2005. pp. 249-250. ISBN: 978-3-901509-46-9. [7] B. Mičieta, M. Gašo, M. Krajčovič, 2014. Innovation performance of organization. In: Communications – Scientific letters of the University of Žilina. Vol. 16, no. 3A (2014), pp. 112-118. ISSN 1335-4205. [8] S. Bangsow, 2010. Manufacturing Simulation with plant simulation and simtalk. Berlin Heidelberg : Springer, 2010. pp. 297. ISBN 978-3642-05073-2. [9] Siemens. 2010. Tecnomatix Plant Simulation 10 Step-by-Step Help. Program manual. pp. 578.