Simulation Study for an Automated Storage and Retrieval System
Anttila, Eero (2017)
Anttila, Eero
2017
Automaatiotekniikka
Teknisten tieteiden tiedekunta - Faculty of Engineering Sciences
This publication is copyrighted. You may download, display and print it for Your own personal use. Commercial use is prohibited.
Hyväksymispäivämäärä
2017-05-03
Julkaisun pysyvä osoite on
https://urn.fi/URN:NBN:fi:tty-201704061256
https://urn.fi/URN:NBN:fi:tty-201704061256
Tiivistelmä
The scope of this work is in simulating an automated storage and retrieval system (AS/RS) and validating that the system works as expected. AS/RSs can be complex systems and development project of an AS/RS can last for several years. Therefore, there is need for a solution that helps the stakeholders to validate that the designed system works as expected. The solution offered in this work is discrete-event simulation of an AS/RS. The simulated example AS/RS consists of stacker cranes moving on aisles, high bay racking, industrial robots, conveyor system, AGV connection, and packing stations. The amount of devices and high capacity requirements make the simulated system complex. The simulated system is under development during the thesis writing process so most of the details about the system functionality are available. However, some uncertainty causing assumptions are required to simulate the AS/RS.
The goal of this work is divided in three research questions. The first research question is concerning the requirements for simulation platform to simulate an AS/RS. The identified requirements are general flexibility and usability requirements; ability to model required devices, warehouse management system, and material flow; and ability to gather the required performance measurements. The second research question deals with the material flow simulations. The material flow simulations present that the example AS/RS can fill the required maximum infeed amounts but outfeed amounts cannot be filled. The reason for this is too low capacity of industrial robots. The third research question is about sensitivity analysis. Firstly, the results present that the realized outfeed amounts are sensitive for variation in robot movement times. If times increase, the realized outfeed amounts decrease significantly. Secondly, the realized infeed amounts are slightly sensitive for total delay of stacker cranes. If delay increases, the realized infeed amount drops a bit. Thirdly, the example AS/RS is not sensitive for variation in storage utilization. This is the benefit of random storage strategy.
Based on the results of this work, the industrial robots are the bottleneck of the example AS/RS. On average, the utilization ratios of industrial robots are 97 % during the maximum infeed and outfeed amount simulations. With better routing of the piles, also the rest of the capacity could be utilized that would lead to higher realized infeed and outfeed amounts. However, then the system would be quite unreliable. Based on the results of this work, the industrial robots require more capacity if designed capacity of the system is required to be filled.
The goal of this work is divided in three research questions. The first research question is concerning the requirements for simulation platform to simulate an AS/RS. The identified requirements are general flexibility and usability requirements; ability to model required devices, warehouse management system, and material flow; and ability to gather the required performance measurements. The second research question deals with the material flow simulations. The material flow simulations present that the example AS/RS can fill the required maximum infeed amounts but outfeed amounts cannot be filled. The reason for this is too low capacity of industrial robots. The third research question is about sensitivity analysis. Firstly, the results present that the realized outfeed amounts are sensitive for variation in robot movement times. If times increase, the realized outfeed amounts decrease significantly. Secondly, the realized infeed amounts are slightly sensitive for total delay of stacker cranes. If delay increases, the realized infeed amount drops a bit. Thirdly, the example AS/RS is not sensitive for variation in storage utilization. This is the benefit of random storage strategy.
Based on the results of this work, the industrial robots are the bottleneck of the example AS/RS. On average, the utilization ratios of industrial robots are 97 % during the maximum infeed and outfeed amount simulations. With better routing of the piles, also the rest of the capacity could be utilized that would lead to higher realized infeed and outfeed amounts. However, then the system would be quite unreliable. Based on the results of this work, the industrial robots require more capacity if designed capacity of the system is required to be filled.