Simulation and Kaizens

Authors Bernard J. Schroer, Mel Adams and Michael McNairy
Center for Automation and Robotics
University of Alabama in Huntsville
Huntsville, Alabama 35899
This paper is being used with permission of the author.

ABSTRACT

This paper presents a case study of using simulation to support a design Kaizen for a proposed manufacturing line. Included in the paper are a brief description of the Kaizen process, the simulation model of the proposed manufacturing line, and the results of using simulation to support the Kaizens.

INTRODUCTION

Lean manufacturing is a systematic approach to identify and eliminate waste, or non-value added activities, through continuous improvement. Competition is forcing manufacturers to improve quality, reduce deliver time and lower cost. Some of the lean tools used to address these concerns are value stream mapping, work place organization (5 S’s), seven wastes, standardized work, setup reduction, total productive maintenance, one-piece flow and cells (Grief, 1991; Imai, 1986; Nakajima, 1998; Rother and Shook, 1999; Shingo, 1986; Shingo, 1989; Sekine, 1990; and UAH 2000).

The key to lean manufacturing is to compress time by eliminating waste and thus continually improving the process. Kaizen is a Japanese word for continuous improvement. A Kaizen event is a focused, intense, team effort aimed at improving a process. The steps in a Kaizen event are given in Figure 1. An event is generally 3-5 days in length.

One tool that has often been overlooked is simulation. This paper presents the use of discrete event simulation in a Kaizen, focusing on a design Kaizen for a new production line.

USE OF SIMULATION IN A KAIZEN EVENT

Figure 2 outlines the steps with simulation added to a Kaizen event. In summary, the sequence of events of the three-day manufacturing line design simulation Kaizen is:

Step 1 Develop preliminary design of line
Step 2 Develop ProcessModel simulation (Day 1)
Step 3 Validation and verification
Step 4 Run simulation baseline (Day 1)
Step 5 Present simulation results to Kaizen team (Steps 5-11 on Days 2 and 3)
Step 6 Brainstorm opportunities for improvement to manufacturing line
Step 7 Prioritize opportunities and select top opportunity
Step 8 Brainstorm suggestions for top opportunity
Step 9 Develop alternative design of manufacturing line
Step 10 Modify model to reflect new alternative
Step 11 Run simulation and compare against prior model
Step 12 Return to Step 5

CASE STUDY

The case study is based on a small manufacturing company employing 100 associates. The company manufacturers various parts for large earth moving equipment. The company was bidding on a large contract and was designing a new manufacturing line for the potential new business.

Before presenting its proposal and before purchasing equipment for the new line, the decision was made to apply the concepts of lean manufacturing to look at ways to eliminate waste in the proposed manufacturing line. Since only a preliminary design existed, simulation offered an excellent method of observing the line in operation, identifying problems and bottlenecks, and evaluating the proposed line before construction and equipment purchases.

The simulation Kaizen team consisted of Company Operations Manager, Company Line Supervisor, Lean Consultant and Simulation Consultant. As the team brainstormed opportunities for improvement and developed alternative designs of the manufacturing line, the simulation consultant modified the model and provided the team with the results based on the new alternative.

ProcessModel (1999) allowed the team to view the actual production of the line, WIP buildup in front of stations and utilization of stations and associates. Figure 3 is the ProcessModel simulation of the final design of the manufacturing line.

Baseline

The simulation results of the initial design of the manufacturing line were:

Several observations from the baseline run were 1) the initial design did not meet the production goal of 1800, 2) the line was overstaffed, and 3) the purchase of some of the parts, such as the heads, may be less than the actual fabrication of the parts. Armed with this information, a number of Kaizens were conducted during the three days. The following paragraphs briefly describe four of the design Kaizens.

Figure 3.  ProcessModel Simulation (click the image for a larger view)

Kaizen 1: Purchase piston heads

The selected opportunity for improvement was to “purchase some of the heads from an outside vendor.” The simulation was used to determine the desired purchase quantity. The simulation results as a function of percent purchased heads were:

Kaizen 2: Remove underutilized associates

The selected opportunity for improvement was to “eliminate the number of underutilized associates on the line.” The simulation was used to determine the number of associates without impacting production. The simulation results as a function of associates, with purchase of 25% heads, were:

Kaizen 3: Reduce cycle times at Paint and Pallet/Ship Stations

The selected opportunity for improvement was to “reduce the cycle times at the Paint and the Pallet/Ship Stations to eliminate a bottleneck.” The simulation results by reducing the cycle time at the Paint Station from 1.25 to 1.00 (20%) and the cycle time at the Pallet/Ship Station from 0.43 to 0.35 (19%) were:

Kaizen 4: Increase utilization of Associate 2

The selected opportunity for improvement was to “reallocate work content to Associate 2.”  The simulation results by assigning the Wash Tube Assembly task to Associate 2 were:

Station utilization and average station WIP for this simulation were:

The above simulation results point to several other opportunities for improving the manufacturing line. Several of these opportunities for Kaizens are:

• Reducing WIP of 309 at Build Rod Assembly Station
• Reducing WIP of 267 at Auto Resistance Stud Weld Station
• Reducing WIP of 180 at Inertia Welder Station
• Reallocating work Associate at Wash Head and Piston Station (currently utilization is 24%)
• Reallocating work for Associate at Pack Heads and Pistons Stations (currently 64%). This may be difficult since the cycle times vary for the heads and the pistons

CONCLUSIONS

In summary, after only four design Kaizens that were completed in three days, the following improvements were accomplished:

• Purchasing 25% of the piston heads increased production from 1524 to 1777 (15% increase). As a result, an additional lathe did not have to be purchased.
• Purchasing 50% of the pistons heads did not increase production above 1777
• Production remained at 1777 after eliminating four associates
• Associates #1,2,3,4 and 5 utilization was 96%, 92%, 100%, 100% and 100% respectively while still maintaining 1777 production. This compares to the utilization for the initial design of 96%, 76%, 86%, 35% and 55%, respectively.
• Overall the line is fairly well balanced with the exception of WIP buildup at the Build Rod Assy, Auto Resistance Stud Weld, and Inertia Welder Stations.
• Overall associate utilization is generally high with the exception of the Wash and Pack Heads and Pistons Stations.

The following general comments are made concerning the use of simulation in a design Kaizen:

• Simulation is an effective tool in supporting a Kaizen event, especially using Process Model
• The simulation design Kaizen was completed in three days, with one day to develop the simulation model and two days using the simulation models to evaluate various alternatives.
• Developing the simulation model in one day was possible because the Company had already developed a detailed layout of the proposed manufacturing line and had the station cycle times.

ACKNOWLEDGEMENTS

This project was funded in part by the Alabama Technology Network and Alabama Industrial Development Training.

REFERENCES

• [Greif, 1991] Greif, M., 1991: The Visual Factory, Productivity Press, Cambridge, MA.
• [Imai, 1986] Imai, M., 1986: Kaizen: The Key to Japanese Competitive Success, Random House, New York.
• [Nakajima, 1988] Nakajima, S., 1988: Introduction to Total Productive Maintenance, Productivity Press, Portland, OR.
• [ProcessModel, 1999] Users Manual, ProcessModel Corp., Provo, UT.
• [Sekine, 1990] Sekine, K., 1990: One-Piece Flow, Productivity Press, Portland, OR.
• [Rother and Shook, 1999] Rother, M., and J. Shook, 1999: Learning to See, The Lean Enterprise Institute, Brookline MA.
• [Shingo, 1986] Shingo, S., 1986: Zero Quality Control: Source Inspection and the Poka Yoke System, Productivity Press, Portland, OR.
• [Shingo, 1989] Shingo, S., 1989: A Study of the Toyota Production System, Productivity Press, Portland, OR.
• [UAH, 2000] Lean Manufacturing Handbook, University of Alabama in Huntsville, Huntsville, AL.

BIOGRAPHIES

Bernard J. Schroer, P.E., Ph.D. – Industrial Engineering, Oklahoma State University. He is currently Associate Vice President of Research at the University of Alabama in Huntsville. His research focuses on simulation and lean system implementation.

Mel Adams, Ph.D. – Strategic Management – University of Tennessee. He is currently Sr. Business Consultant for the University of Alabama in Huntsville center of the Alabama Technology Network. He is a NIST-MEP certified lean trainer, an experienced kaizen facilitator, and the NIST-MEP Practitioner of the Year for 2000. Research interests focus on strategic planning and lean leadership.

Michael McNairy - B.S. Industrial Engineering - University of Alabama in Huntsville. He is a Research Associate for the University of Alabama in Huntsville Center of the Alabama Technology network. He is a NIST certified lean trainer.