Normative perspectives in manufacturing

Al Kindi, Mahmood A.

Permalink

https://hdl.handle.net/2142/16899 Copy

Description

Title

Normative perspectives in manufacturing

Author(s)

Al Kindi, Mahmood A.

Issue Date

2010-08-20T18:01:13Z

Director of Research (if dissertation) or Advisor (if thesis)

Abbas, Ali E.

Doctoral Committee Chair(s)

Abbas, Ali E.

Committee Member(s)

• Thurston, Deborah L.
• Stipanović, Dušan M.
• Uddin, Rizwan

Department of Study

Industrial&Enterprise Sys Eng

Discipline

Systems & Entrepreneurial Engr

Degree Granting Institution

University of Illinois at Urbana-Champaign

Degree Name

Ph.D.

Degree Level

Dissertation

Keyword(s)

• Decision Analysis
• Six Sigma
• Testing in Product Development
• Multi-attribute

Abstract

This research proposes a normative approach in two infrastructural manufacturing decisions: quality and product development. Quality, as employed in the research, is defined as satisfying customer needs given a market segment. Product defects lead to customer dissatisfaction and have a negative economic impact. These defects can be due to a manufacturing process variation or an ill-advised product design. The former cause of defects is eliminated using Six Sigma. Six Sigma has spread widely due to its successful implementation by GE, Motorola, and other companies. However, there is no notion of using a normative approach in Six Sigma. The second source of product defects is errors in product design that can be reduced by conducting early testing in the product development phase. The normative approach in this research consists of three decision bases: (1) alternatives, (2) preferences, and (3) information. This research raises the questions: when is improving quality of the product worthwhile, and is “process perfection” a wise economic objective? This research builds a decision analytic model for the decision to incorporate a Six Sigma quality process. The objective is to evaluate the economic impact of Six Sigma based on a rigorous approach, given that previous research revolved around best practice. The model employs quality as a binary outcome (good or bad) in order to examine the effects of some key elements regarding the Six Sigma decision. These elements include implementation cost, firm size, defect costs, and defects’ opportunities (number of production or service stages). The optimal solution reveals several managerial insights regarding the impact of the various factors related to the Six Sigma decision. Naturally, implementation cost makes Six Sigma less attractive, while attitude toward risk plays a role in determining the optimal sigma (quality) level. In addition, the best decision alternative for small- and middle-sized firms is not necessarily to have the highest quality standard or the highest sigma level. In some instances it may be more important to consider other profit-generating alternatives before taking the Six Sigma route. The market characteristic (demand –price relation) and competitors’ performance and price provide significant impact with regard to the economical value of Six Sigma. The firm values its process perfection more when the competitor is inferior in price and performance. Moreover, it is not necessary to improve every manufacturing process to achieve economical benefits. Timing and frequency of reliability tests during the product development phase is essential for eliminating design errors. Testing is modeled as an activity that generates information about technical errors and problems related to customer needs that will require redesign. Optimal testing strategies (number and timing of tests) must balance the tradeoff among several variables, including the cost of a test, the increasing cost of redesign when discovered at a later stage, and the relationship between sequential tests. This research investigated two testing environments: deterministic and stochastic. Optimal strategies in both domains are obtained. The strategy reveals several managerial insights. The design team should run more tests when the mathematical relationship between the redesign cost and the accumulated error is linear when compared to concave or convex relationships. Attitude toward risk has a major role regarding the optimal number of tests done within a stochastic domain. The nature of redesign cost has the most effect on risk aversion when compared to neutral or risk seeking behaviors. Moreover, the team ought to run more tests when the redesign cost is a function of time elapsed between tests, compared solely to clock time. Learning by doing increases the optimal number of tests.

Graduation Semester

2010-8

Permalink

http://hdl.handle.net/2142/16899

Copyright and License Information

Copyright 2010 Mahmood A. Al Kindi

Owning Collections