Type of Document Dissertation Author Diponegoro, Ahmad URN etd-0702103-012837 Title Finite-Horizon Operations Planning for a Lean Supply Chain system Degree Doctor of Philosophy (Ph.D.) Department Engineering Science (Interdepartmental Program) Advisory Committee
Advisor Name Title Bhaba R. Sarker Committee Chair Evangelos Triantaphyllou Committee Member Jerry L. Trahan Committee Member Lawrence Mann, Jr. Committee Member Sitharama S. Iyengar Committee Member Sudipta Sarangi Dean's Representative Keywords
- network optimization
- time-varying demand
Date of Defense 2003-05-13 Availability unrestricted AbstractThis dissertation studies an operational policy for a lean supply chain system consisting of a manufacturer, multiple suppliers and multiple buyers. The manufacturer procures raw materials from the suppliers and converts them into finished products, which are then shipped in batches to the buyers at certain intervals of times. Three distinct but inseparable problems are addressed: single supplier and single buyer with fixed delivery size (FD), multiple suppliers and multiple buyers with individual delivery schedule (MD), and time dependent delivery quantity with trend demand (TD). The mathematical formulations of these supply systems are categorized as mixed-integer, nonlinear programming problems (MINLAP) with discrete, non-convex objective functions and constraints.
The operations policy determines the number of orders of raw material, beginning and ending times of cycles, production batch size, production start time, and beginning and ending inventories. The goal is to minimize the cost of the two-stage, just-in-time inventory system that integrates raw materials ordering and finished goods production system. The policy is designed for a finite planning horizon with various phases of life cycle demands such as inception (increasing), maturity (level) and phasing out (declining).
Analytical results that characterize the exact, optimal policy for the problems described above are devised to develop efficient and optimal computational procedures. A closed-form heuristic that provides a near-optimal solution and tight lower bound is proposed for the problem FD. A network model to represent the problems is proposed and network-based algorithms are implemented to solve the problems FD, MD and TD optimally. The computational complexities of the algorithms are Θ(N2) or O(N3) where N is the total number of shipments in the planning horizon. Numerical tests to assess the robustness and quality of the methods show that the present research provides superior results.
Production and supply chain management play an important role in ensuring that the necessary amounts of materials and parts arrive at the appropriate time and place. A manager, using the models obtained in this research, can quickly respond to consumers' demand by effectively determining the right policies to order raw materials, to deliver finished goods, and to efficiently manage their production schedule.
Filename Size Approximate Download Time (Hours:Minutes:Seconds)
28.8 Modem 56K Modem ISDN (64 Kb) ISDN (128 Kb) Higher-speed Access Diponegoro_dis.pdf 2.79 Mb 00:12:53 00:06:37 00:05:48 00:02:54 00:00:14
If you have questions or technical problems, please Contact LSU-ETD Support.