Analysis of Production Layout Alternatives Using Lean Techniques and Monte Carlo Simulation

Transcription

1 Analysis of Production Layout Alternatives Using Lean Techniques and Monte Carlo Simulation David Henry, David Shindledecker and Faisal Aqlan* Master of Manufacturing Management (MMM) Program Penn State Erie, The Behrend College Erie, PA 16563, USA Abstract With business pressures on the rise, it is more important than ever to stay competitive and develop unique ways to reduce costs. Companies should focus on minimizing non-value added work to decrease indirect material costs and improve material movement throughout the plant. In this paper, we present a study to analyze production layouts in a manufacturing company utilizing Lean techniques and Monte Carlo simulation. We consider three layout alternatives for a stockroom with the objective to reduce material movement and non-value added work as well as improve safety by reducing forklift traffic. Results from the case study show that the current location of the stockroom is the optimal when considering distance of travel and cost of moving the material. Keywords Production layout, manufacturing industry, Lean, Monte Carlo Simulation. 1. Introduction With business pressures on the rise, it is more important than ever to stay competitive and come up with unique ways to reduce costs. Companies should focus on minimizing non-value added work to decrease indirect material cost and improve material movement throughout the plant. The objective of this study is to reduce the material movement, increase material flow and improve material presentation within a local manufacturing company. The production line of interest currently feeds two main final assembly areas. These final assembly areas also are fed by 5 product focused sub-assembly areas. All of the assembly is currently done in Bay A on the east side of the building. The material supermarket that feeds material to these lines is located on the opposite side of the building in Bay E, and the receiving area is in the northwest corner of the building and feeds all of the aforementioned areas. The layout of the production line is shown in Figure 1. Figure 1. Production line layout 1154

2 With recent changes to the building an opportunity has arose to possibly move the stockrooms for the whole building to the 2 nd floor, which is located close to the receiving dock in an attempt to improve material flow and reduce material movement in the building. This study will analyze what the change would do to the flow of materials and if the cost and hassle of moving the stockroom would be worth the benefit that we would get by reducing the non-value added work of moving material around the building. If the movement of the stockroom proves to be a good improvement for material flow, the plan is move all building stockrooms to this centrally located one. With the indefinite possibility of the supermarket growing, we want to have avid space as we have future plans to introduce material kitting to all bays individually. With these continuous improvements, we are introducing new ways to reduce or eliminate non-value added work and using Lean principles and Monte Carlo simulation. These changes not only will reduce material movement and non-value added work, but they will increase safety by reducing forklift traffic. It is improvements like these that keep the company competitive in the marketplace and changing to adapt to the new economy to stay relevant for years to come. To understand how material moves throughout the plant, we first looked at the process by creating a process flow diagram (PFD) for receiving, stocking and kitting (see Figure 2). The first diagram describes the receiving process from when a truck arrives at the shipping dock with material, to when that material is delivered to the stockroom. The second PFD is that of the receiving process which shows what the material handlers due after the parts show up in the supermarket. Last is the kitting process, where the material handlers organize the parts on a kit cart, to make it easy to see material shortages and to organize the material for the assemblers to be more efficient. Figure 2. Process flow diagram for receiving, stocking, and shipping processes 1155

3 2. Relevant Literature Process improvement methodologies are widely used in manufacturing and service industries to reduce waste ad variation and improve productivity. Aqlan (2018) presented a study on implementing Lean six sigma process improvement in casting industry. Several process improvement projects were identified and five projects were selected to improve the productivity of the foundry. Several studies in the literature have discussed optimization of production layout for different reasons. For instance, Thottungal (2008) have created and Automated Layout Design Program (ALDEP) to simulate the best form of layout that is required. The study argues that the optimal layout strategy for a company is a mix of product line layout and process layout. If the work stations are not located in such a manner as to facilitate straight line flow throughout the facility, the back and forth movement of the product will add a significant amount to the time and cost. A relationship table was used to determine the value of movements between various departments. Vishnu et al. (2017) have conducted a study to improve the plant efficiency and machine utilization by finding the most efficient arrangement of each workstation. The study was performed by Computerized Relative Allocation of Facility technique. This technique allows to figure out which station was or was not in line with production process. For this process, each department was measured by its area and its centroids were identified by different colors. As discussed in Hung and Maleki (2013), Group Technology (GT) technique can result in shorter development and production lead time. However, GT implementation could be challenging based on the correct grouping of parts into families. Some coding system such as hierarchical, chained-type and hybrid structures are needed to achieve this process. Even though, GT is often appropriate in metal machining operation, still, the coding systems may not be universally applicable. The authors presented a case for applying GT in forging industry. Simulation and optimization methods can also be used to optimize production layouts. Aqlan et al. (2014) discussed the use of simulation and optimization techniques to improve production layout in server manufacturing by consolidating production line and changing product layout to process layout. In this paper, we analyze production layouts in a manufacturing company utilizing Lean manufacturing techniques and Monte Carlo simulation. By optimizing the production layout, the total transportation cost can be reduced. 4. Research Methodology In this study, we consider three different locations for the stockroom, including the current location. We analyze the three alternatives based on distance and cost of movement. The following sections discuss the details of the research methodology Layout Options We have two alternative layout options that we are going to analyze to see if they would be effective in reducing material movement and reduce costs associated with the material movement. In Figure 3, we show the current layout of the plant along with the daily movement of material within the plant. As you can see in Figure 4, the current stockroom is in the center of the west side of the building. The assembly lines, in green, are located on the East side of the building. The shipping dock, in blue, where material is received in is in the North West corner of the building. With the current option, the shipping dock employees have to come half way across the building as you can see in Figure 4 in red. Then the stockroom material handlers have to take the material from the west side of the building to the east side of the building. 1156

4 Figgure 3. to Stockroom locations (daily deliveries) The first option, in yellow, is to move the stockroom to the front of the building on the second floor. The benefits that we see, is that when material comes in it is a very short distance to stock the material in the stockroom, because its minimal distance from the shipping dock. Also, another benefit of this option is that currently the 2 nd floor has been emptied out and would be an easy solution to move to due to the fact that you wouldn t have to disrupt an already occupied area. The downfalls to this solution, is that it would now add elevator moves to the material movement. There are three elevators, a material only conveyor, a passenger elevator, and a freight elevator that is also able to transfer a small jitney. The elevators do break down on occasion, but normally never at the same time, which would have to be considered if we go this route. The second option is to move the stockroom closer to the assembly lines, in Bay B. This option is beneficial because it reduces the movement from the stockroom to the lines, which is a main driver in the jitney traffic. Unfortunately, there is currently an assembly workstation in this area, so if we do end up finding great benefit to moving to this location, we would also have to factor in the cost of moving the equipment out of the area first, so we could move into the stockroom. Figure 4. The options for moving stockroom location the material flow paths 1157

5 So, as we can see there are three different route options for jitney traffic, depending on where the stockroom is located, all with different benefits and different weaknesses. In order to figure out which would be the best one financially, we are going to consider multiple factors in a cost analysis to see the financial impact of each Calculation of Total Distance In order to evaluate the layout alternatives, we conduct physical measurement of the distances between the stockroom location and the assembly lines as well as analyzed historical data to calculate the frequency of moving the material. The total distance is calculated as Total Distance = Distance * Frequency. Figure 5 shows an example for calculating the total distance. Current Location Current Location 6E15 6E16 6E17 6E18 6Y02 6Y03 6Y E E16 0 6E17 0 6E18 0 6Y Y03 0 6Y04 0 6Y05 0 6A A A A A B11 0 6B12 0 6B13 0 6Y05 6A04 6A06 6A15 6A20 6A21 6B11 6B12 6B13 * 6E15 6E16 6E17 6E18 6Y02 6Y E E16 0 6E17 0 6E18 0 6Y02 0 6Y03 0 6Y04 0 6Y05 0 6A04 0 6A06 0 6A15 0 6A20 0 6A21 0 6B11 0 6B12 0 6B13 0 6Y04 6Y05 6A04 6A06 6A15 6A20 6A21 6B11 6B12 6B13 Current Location = 6E15 6E16 6E17 6E18 6Y02 6Y03 6Y E E E E Y Y Y Y A A A A A B B B Y05 6A04 6A06 6A15 6A20 6A21 6B11 6B12 6B Material Movement Cost Analysis Figure 5. An example for calculating the total distance A major factor on deciding to relocate the stockroom to one of the optional locations is the average cost for each movement of material. To analyze the cost of material movement, we used the overall shop labor rate in the calculations. We collected the data from the shipping receipts (production receipts) to identify the number of shipments that were delivered to each stockroom on a monthly average. The distances were also calculated (measured in feet) from each stockroom location to the receiving shipping dock location. The distances were then 1158

6 Proceedings of the International Conference on Industrial Engineering and Operations Management converted to time based on the average fork truck speed of 4 miles per hour. The average delivery cost was then calculated (Table 1) to be $5.54 per delivery. Table 1. An example of calculating the material movement cost 4.4. Monte Carlo Simulation Monte Carlo simulation is used to model the probability of different outcomes in a process that cannot easily be predicted due to intervention of random variables. In this particular case, we used Crystal Ball ( software to obtain the probabilities of how many material moves would be down from the shipping dock to the stockroom and the stock room to the assembly areas. With each of these with a probability of how many times a month they would have to do a material move, we run the Monte Carlo simulation for 1,000 trials. As we can see in Figure 6, the cost of material movements for the current location was in the middle of both of the other options and averaged about $1,200 per month to move material, the 2nd floor option was unfortunately about $300 per month more expensive than the current location, so it would not be a viable option. However, the option to move it to Bay B did decrease costs associated from moving material from the current location. This saves roughly $200 per month on average, which would be a savings of $2,400 per year. Again, what we have to factor into this option is that we would have to move equipment out of the area before we would be able to move the stockroom into the area to generate the savings. 1159

7 4.5. Cost of Moving the Equipment Figure 6. Monte Carlo Simulation for costs of material movements Another factor in the analysis to move stockroom locations is the cost to free up the shop floor space by moving existing equipment. The costs are calculated based on internal maintenance personnel shop labor rates and existing Environmental Health & Safety (EHS) standards. The moves for each stockroom can then be calculated as shown in Table 2. The costs in Table 2 are the estimated costs associated with the relocation of equipment from the proposed option 2. These estimates are based on type and quantity of equipment and they have been applied to all of the proposed stockroom relocations in this study. Table 2. Costs by equipment type for layout option 2 Item Cost per unit Quantity Total Cost Ridge-U-Rack shelving units $ $6, Small Equipment (No oil sump) $ $6, Small Equipment (oil sump) $ $ Electrical Drops $ $5, Note: Equipment must be newer than 1990, or additional testing ($200. per unit) needed for $18, PCB's. 5. Summary and Final Recommendations When starting this study, we and several of the leadership team were convinced that the 2 nd floor stockroom option would be a sure winner. After conducting the detailed analysis, the results does not support the original hypothesis. The best recommendation is to stick with the existing stockroom location. The other two options have a negative payback once the costs of moving equipment are factored in. The delivery cost savings are just not significant enough to offset the difference of moving equipment. Maybe in the future, the costs of a project like this will be valued at different rates (1 st floor real-estate for production has a higher value) that may make this project a viable option. The biggest takeaway would be to always perform the proper analysis for a project like this, the data may say something different than your gut feeling does. 1160

8 References Thottungal, A.P., Redesign the layout of a forging unit using discrete event simulation, International Journal of Emerging Technology and Advanced Engineering, vol. 3, no. 8, pp. 1-7, Aqlan, F., Using Lean six sigma methodologies to improve operational performance in foundries, International Journal of Lean Enterprise Research, 2018, in press. Vishu, N.A., Rakesh, P.R., and Surendran, A., Optimization of manufacturing plant layout design in SIFL using CRAFT, International Journal of Current engineering and Scientific Research, vol. 4, no. 7, pp , Aqlan, F., Lam, S.S., Ramakrishnan, S., An integrated simulation-optimization study for consolidating production lines in a configure-to-order environment, International Journal of Production Economics, vol. 148, pp , Hung, K.T., and Maleki, H., Applying group technology to the forging industry, Production Planning and Control, vol. 25, no. 2, pp , Biographies David Henry, David Shindledecker are master students in the Master of Manufacturing Management (MMM) program at Penn State Erie, The Behrend College. Faisal Aqlan is currently an assistant professor of Industrial Engineering and Master of Manufacturing Management (MMM) at Penn State Behrend. He earned his Ph.D. in Industrial and Systems Engineering from the State University of New York at Binghamton in Aqlan has worked on industry projects with Innovation Associates Company and IBM Corporation. His work has resulted in both business value and intellectual property. He is a certified Lean Silver and Six Sigma Black Belt. He is a senior member of the Institute of Industrial and Systems Engineers (IISE) and currently serves as the president of IISE Logistics and Supply Chain Division, director of Young Professionals Group, and founding director of Modeling and Simulation Division. Aqlan is also a member of American Society for Quality (ASQ), Society of Manufacturing Engineers (SME), and Industrial Engineering and Operations Management (IEOM) Society. He has received numerous awards including the IBM Vice President award for innovation excellence, Penn State Behrend s School of Engineering Distinguished Award for Excellence in Research, and the Penn State Behrend s Council of Fellows Faculty Research Award. Aqlan is the Principal Investigator and Director of the NSF RET Site in Manufacturing Simulation and Automation at Penn State Behrend. 1161