|
A MULTI-ITEM TWO-ECHELON INVENTORY PROBLEM UNDER JOINT REPLENISHMENT POLICY |
|---|---|
| รหัสดีโอไอ | |
| Title | A MULTI-ITEM TWO-ECHELON INVENTORY PROBLEM UNDER JOINT REPLENISHMENT POLICY |
| Creator | Varaporn Pukcarnon |
| Contributor | Paveena Chaovalitwongse, Naragain Phumchusri |
| Publisher | Chulalongkorn University |
| Publication Year | 2556 |
| Keyword | Inventory control, Inventory control -- Simulation methods, Heuristic algorithms, การควบคุมสินค้าคงคลัง, การควบคุมสินค้าคงคลัง -- การจำลองระบบ, ฮิวริสติกอัลกอริทึม, ปริญญาดุษฎีบัณฑิต |
| Abstract | This dissertation studies a multi-item two-echelon inventory problem under a joint replenishment policy called “the can-order policy”. The system is composed of one warehouse and multiple retailers facing stochastic demand, and all locations are replenished continuously. This research considers lead time and target service level as system constraints. The research is conducted in three phases: phase I – a single-item system with zero lead time, phase II – a single-item system with non-zero lead time, and phase III – a multi-item system with non-zero lead time. Each phase contains different number of decision variables and relevant factors. Due to the system complications, computer simulation is initially utilized for inventory policy setting. It provides insights of inventory policy setting: the effects of relevant factors and the solution characteristics. Heuristic approaches are developed to solve the problem for each phase. The proposed heuristics are based on decomposition approach, iterative procedure, and one-dimensional search called golden section search to determine the appropriate inventory policy setting. For phase I and II, the proposed heuristics’ performance is measured against the best-known solution providing the minimum average total system-wide cost. The best-known solution can be determined by computer simulation with systematic procedures: input determination and output validation. From the experimental results, the proposed heuristics can obtain the appropriate policy much faster than computer simulation with the average cost gap at 1.54% for phase I and 1.20% for phase II, respectively. For phase III, this research provides comparative analysis of the proposed heuristics to identify which situation is suitable for each heuristic. |
| URL Website | cuir.car.chula.ac.th |
