검색 상세

Strategic WIPI Positioning in Demand Driven MRP

초록/요약

Now more than ever a decisive competitive edge can be achieved by companies with a high degree of service level. It becomes so critical to meet the lead time that customers require. Therefore, in most make-to-order manufacturing, it is necessary to hold work-in-process inventory (WIPI) at almost every station. However, some of the WIPI in the network do not contribute to reducing the manufacturing lead time of the final product. Hence, the first question of effective WIPI management is “where to place decoupling inventory in the station network”, called strategic inventory positioning, instead of “how much inventory should have” and “when to place order”. In this paper, we develop a framework for modeling the optimal set of stations to hold WIPI so that the total inventory holding cost is minimized, while proving a high level of the service to the customer depending on meeting the required due date of the final product. Because of the inaccuracy forecasting and uncertainty in operations management and processes, MRP may result in excessive WIPIs. So Carol Ptak and Chad Smith proposed “Demand Driven MRP” in “Orilicky’s MRP” 3rd edition in 2011, which shows a new formal planning and execution method. They introduced new contents including Actively Synchronized Replenishment (ASR) lead time and buffer profile that would be applied in this paper. The paper proposes four models to determine the optimal position and quantity of WIPI for different given bill of materials (BOMs) tree structure of products with/ without processing machine constraints. We consider multi- level BOMs in this paper as the simple bill of material (S-BOM), in which parent-child in the BOM has a one-to-many relationship, and the general BOM (G-BOM), in which parent-child in the BOM has a many-to-many relationship. In the BOM structure, the same machine may be used to process different subcomponents, called processing machine constraints. The machine scheduling problem needs to be considered while there are processing machine constraints. In this paper, four models are discussed for two kinds of given BOMs with/without processing machine constraints including SIP-S problem (strategic inventory positioning problem with a simple BOM without machine constraints), SIP-G (SIP problem with a general BOM without machine constraints), SIP-S-C (SIP problem with a simple BOM with machine constraints) and SIP-G-C (SIP problem with a general BOM with machine constraints) The paper builds four mathematical models based on four situations. However, these models are non-linear programming (NLP) models so that it is hard to solve. Therefore, genetic algorithms are developed to solve these models.

more

목차

Chapter 1. Introduction 1
1.1 Background and Motivation 1
1.2 Scope and Objective 4
1.3 Outline 5
Chapter 2. Literature Review 7
Chapter 3. Demand Driven MRP 10
3.1 ASRLT in S-BOM, G-BOM with/without Processing Machine Constraints 10
3.1.1 ASTLT in S-BOM without Processing Machine Constraints 11
3.1.2 ASTLT in G-BOM without Processing Machine Constraints 11
3.1.3 ASTLT in S-BOM with Processing Machine Constraints 15
3.1.4 ASTLT in G-BOM with Processing Machine Constraints 15
3.2 Buffer Profiles 17
Chapter 4. Mathematical Models for SIP problem. 21
4.1 SIP Problem in S-BOM without Processing Machine Constraints (SIP-S Problem) 22
4.2 SIP Problem in G-BOM without Processing Machine Constraints (SIP-G Problem) 26
4.3 SIP Problem in S-BOM with Processing Machine Constraints (SIP-S-C Problem) 32
4.4 SIP Problem in G-BOM with Processing Machine Constraints (SIP-G-C Problem) 36
Chapter 5. Genetic Algorithm Based Solution Procedure for 4 models 39
5.1 GA Based Solution Procedure for SIP-S Problem 40
5.1.1 GA Based Solution Procedure for SIP-S Problem 40
5.1.2 Numerical examples for SIP-S problem 42
5.2 GA Based Solution Procedure for SIP-G Problem 47
5.2.1 GA Based Solution Procedure for SIP-G Problem 47
5.2.2 Numerical examples for SIP-G problem 47
5.3 GA Based Solution Procedure SIP-S-C and SIP-G-C Problems 52
Chapter 6. Conclusion 53
References 54

more
반출 Meta View 목록