DOC PREVIEW
MIT 15 763J - Use of a Queueing Model to Design a Lean System

This preview shows page 1 out of 4 pages.

Save
View full document
View full document
Premium Document
Do you want full access? Go Premium and unlock all 4 pages.
Access to all documents
Download any document
Ad free experience
Premium Document
Do you want full access? Go Premium and unlock all 4 pages.
Access to all documents
Download any document
Ad free experience

Unformatted text preview:

Massachusetts Institute of Technology Leaders for Manufacturing Program Use of a Queueing Model to Design a Lean System Prepared as a basis for class discussion by Jamie Flinchbaugh1. Toyota has evolved the design of their vehicle assembly plants for 50 years. Much of the work coincided with the development of the Toyota Production System2 and has evolved through periods of trial and error. In this case, with the corresponding model found in lean_factory.xls, you have the opportunity to use a queueing model to explore the relationship that Toyota discovered when developing the Toyota Production System. A brief explanation of the factory dynamics In a factory using the andon process3, buffers are used to create independence among departments attempting to push decision making further down in the hierarchy and create mini-companies within the factory. This is attained by physically promoting significantly more independent departments or line segments than a traditional assembly plant would have, as depicted in the figure. A line segment may consist of somewhere between 20-40 workstations. Separating each line segment is an accumulating buffer that can hold several work cycles of product. The buffers allow each of the teams to make decisions regarding stopping the line to fix problems. The buffers also increase the independence in operating metrics. These buffers seem to violate the principle that inventory is waste and should be eliminated; instead these buffers de-couple one line segment from another, and thus prevent downtime at one segment from shutting down downstream segments on account of no inventory. 1 Jamie Flinchbaugh is a Fellow in the Leaders for Manufacturing Program, Class of 1998, and is sponsored by Chrysler Corporation. This case is extracted from Implementing Lean Manufacturing Through Factory Design,an MIT Thesis by Jamie Flinchbaugh, 1998. 2 This case requires a working knowledge of the Toyota Production System. An adequate understanding can be reached by reviewing Harvard Business School case study Toyota Motor Manufacturing, U.S.A., Inc. (case number 9-693-019, September 5, 1995, prepared by Kazuhiro Mishina). 3 The andon process: each line worker can signal for help from the team leader if there is a quality problem. The team leader makes a decision on whether or not to stop the line, but if at all possible, will make sure that the quality problem is fixed before leaving the workstation. This results in problem solving closer to the problem and very little rework, resulting in a better quality product.As far as managing the line, one group leader leads the team in each line segment. Under the group leader may be three or four team leaders who support teams of 6-10 team members. Both team members and the team leader are union jobs. The group leader has much more broad responsibility than in a traditional assembly plant, to the extent that they act as the president of a mini-company, with the upstream line segment as the supplier and the downstream line segment as the customer. What about the factory designer? The factory designer must relate this operational understanding into the physical design parameters of a factory. The designer must balance investment cost decisions with the operating performance of the factory such as cost, quality, and throughput. You, as factory designer, must be concerned with the relationships between the operation of the andon system (fix quality in-station) and throughput losses. If a series of processes or workstations have equal capacity, but their production varies (either within cycle times or through downtime), the output will never equal capacity. Said another way, the system’s realized capacity, or system throughput, will always be lower than the capacity of the process with the minimum stand-alone capacity. In order to get the desired capacity, there are three compensating techniques that can be used independently or in combination:  Reduce processing variation  Provide excess capacity so that realized throughput equals customer demand  Provide decoupling buffers between processes to reduce the impact adjacent processes have on each other Each of these solutions has other factors or trade-offs to consider. Providing buffers increases system lead-time, which in turn increases work-in-process inventory, hurts problem solving capabilities, and allows more opportunities for in-system damage. Providing excess capacity could be restated as allowing for lower plant efficiency. Excess capacity is costly both to investment and variable costs. Reducing processing variation, which would mean both variance within cycle times and improving equipment uptime, is perhaps the least costly, but requires significant and specific skills within the organization. Although Toyota uses all three solution points to some degree, reducing processing variation is their primary focus. The three compensating techniques are not a complete factory design, but they do compose the strategic factory design. While these design factors exist for any factory, we need to relate them to the design of a vehicle assembly plant. Achieving strategic goals relating to variation, capacity, and buffers will be achieved through management of the physical design features described in the following table.Table of design features Physical design parameters What they represent Base case values Number of line segments The total number of workstations is broken up into line segments. The size of the line segments is determined by how many pieces the total is broken up into, which is the number of buffers plus one. This value does not show up explicitly in the model. It is represented in the length of the line segments. It should be considered, however, because each new line segment represents significant cost. The base case represented six line segments. Length of line segments (or # stations / segment) The length of the line segments is the total number of workstations, typically 250-500 for an assembly plant, divided by the number of line segments. We settled on a value of 33, which is in the ballpark of Toyota’s factories. If there were no buffers, i.e. a traditional assembly plant, this value would reach 200-500. Size of buffers (or accumulator size) The size of the buffers represents how many vehicles can be held in each buffer between line segments. We settled on a value of 15. Buffer size drives conveyor length


View Full Document

MIT 15 763J - Use of a Queueing Model to Design a Lean System

Download Use of a Queueing Model to Design a Lean System
Our administrator received your request to download this document. We will send you the file to your email shortly.
Loading Unlocking...
Login

Join to view Use of a Queueing Model to Design a Lean System and access 3M+ class-specific study document.

or
We will never post anything without your permission.
Don't have an account?
Sign Up

Join to view Use of a Queueing Model to Design a Lean System 2 2 and access 3M+ class-specific study document.

or

By creating an account you agree to our Privacy Policy and Terms Of Use

Already a member?