Execution in supply chain management : dynamics, missteps and mitigation strategies / Noel Watson.

Format:

Book

Manuscript

Microformat

Thesis/Dissertation

Author/Creator:

Watson, Noel.

Contributor:

Zheng, Yu-Sheng, advisor.

University of Pennsylvania.

Language:

English

Subjects (All):

Penn dissertations--Operations and information management.

Operations and information management--Penn dissertations.

Penn dissertations--Managerial science and applied economics.

Managerial science and applied economics--Penn dissertations.

Local Subjects:

Penn dissertations--Operations and information management.

Operations and information management--Penn dissertations.

Penn dissertations--Managerial science and applied economics.

Managerial science and applied economics--Penn dissertations.

Physical Description:

xxvi, 219 pages ; 29 cm

Production:

2002.

Summary:

Given common observations of poor supply chain performance, we examine and highlight possible dynamics at work in supply chain inventory management that cause poor supply chain execution. We also examine strategies with potential for mitigating the effects of these dynamics.

The supply chain dynamics that we consider in this dissertation are inventory management errors as a result of managers over-reacting to demand changes or using textbook forecasting techniques inappropriately, the focus on order fulfillment within the chain and the attendant error compounding as we move up the chain as in the Beer Game, order generation leadtimes and natural order variability. The strategies or recommendations examined are: conservatism in forecasts and use of textbook techniques, chain management focused on end-customer demand, demand information sharing and deliberate order variability mitigation. In particular, we examine decentralization/coordination schemes that allow upstream chain members to correct for downstream mistakes.

Our work is facilitated primarily by the study of inventory management given demand is AR(1), a simple serially correlated demand process. The demand process is the basis for modeling an over-reacting manager and provides the basis for studying natural order variation. Further, though we will initially study dynamics such as order generation leadtimes and the focus on order fulfillment within the chain using i.i.d. demand, some of the issues surrounding these dynamics are more visible when demand is AR(1).

The major contribution of this dissertation is our demand focused decentralization scheme for serial supply chains, facilitated by echelon inventory management and creative accounting. This scheme allows a central manager to reduce the error compounding associated with order focused schemes considerably and given a heuristic for determining the parameters of the incentive scheme, provides a model for decentralized decision making within a vertically integrated chain. This decentralized decision making model has the features that the incentive scheme parameters are demand distribution free and further that the owner does not need to know how to determine the optimal policy for the system.

Notes:

Adviser: Yu-Sheng Zheng.

Thesis (Ph.D. in Operations and Information Management) -- University of Pennsylvania, 2002.

Includes bibliographical references.

Local Notes:

University Microfilms order no.: 3073069.

OCLC:

244972893