Date of Award
8-2022
Document Type
Dissertation
Degree Name
Doctor of Philosophy (PhD)
Department
Industrial Engineering
Committee Chair/Advisor
Dr. Tugce Isik
Committee Member
Dr. Mary E. Kurz
Committee Member
Dr. Scott J. Mason
Committee Member
Dr. Yongjia Song
Abstract
A manufacturing company's inbound supply chain consists of various processes such as procurement, consolidation, and warehousing. Each of these processes is the focus of a different chapter in this dissertation.
The manufacturer depends on its suppliers to provide the raw materials and parts required to manufacture a finished product. These suppliers can be located locally or overseas with respect to the manufacturer's geographic location. The ordering and transportation lead times are shorter if the supplier is located locally. Just In Time (JIT) or Just In Sequence (JIS) inventory management methods could be practiced by the manufacturer to procure the raw materials and parts from the local supplier and control the inventory levels in the warehouse. In contrast, the lead time for the orders placed with an overseas supplier is usually long because sea-freight is often used as a primary mode of transportation. Therefore, the orders for the raw materials and parts (henceforth, we collectively refer to raw material and part by part) procured from overseas suppliers are usually placed using forecasted order quantities. In Chapter 2, we study the procurement process to reduce the overall expected cost and determine the optimal order quantities as well as the mode of transportation for procurement under forecast and inventory uncertainty. We formulate a two-stage stochastic integer programming model and solve it using the progressive hedging algorithm, a scenario-based decomposition method.
Generally, the orders are placed with overseas suppliers using weekly or monthly forecasted demands, and the ordered part is delivered using sea-containers since sea-freight is the primary mode of transportation. However, the end manufacturing warehouse is usually designed to hold around one to two days of parts. To replenish the inventory levels, the manufacturer considered in this research unloads the sea-container that contains the part that needs to be restocked entirely. This may cause over-utilization of the manufacturer's warehouse if an entire week's supply of part is consolidated into a single sea-container. This problem is further aggravated if the manufacturer procures hundreds of different parts from overseas suppliers and stores them in its warehouse. In Chapter 3, we study the time-series forecasting models that help predict the manufacturing company's daily demand quantities for parts with different characteristics. The manufacturer can use these forecasted daily demand quantities to consolidate the sea-containers instead of the weekly forecasted demand. In most cases, there is some discrepancy between the predicted and actual demands for parts, due to which the manufacturer can either have excess inventory or shortages. While excess inventory leads to higher inventory holding costs and warehouse utilization, shortages can result in substantially undesirable consequences, such as the total shutdown of production lines. Therefore, to avoid shortages, the manufacturer maintains predetermined safety stock levels of parts with the suppliers to fulfill the demands arising from shortages. We formulate a chance-constraint optimization model and solve it using the sample approximation approach to determine the daily safety stock levels at the supplier warehouse under forecast error uncertainty.
Once the orders are placed with the local and overseas suppliers, they are consolidated into trailers (for local suppliers) and sea-containers (for overseas suppliers). The consolidated trailers and sea-containers are then delivered to the manufacturing plant, where they are stored in the yard until they are called upon for unloading. A detention penalty is incurred on a daily basis for holding a trailer or sea-container. Consolidating orders from different suppliers helps maximize trailer and sea-container space utilization and reduce transportation costs. Therefore, every sea-container and trailer potentially holds a mixture of parts. When a manufacturer needs to replenish the stocks of a given part, the entire sea-container or trailer that contains the required part is unloaded. Thus, some parts that are not imminently needed for production are also unloaded and stored inside the manufacturing warehouse along with the required parts. In Chapter 4, we study a multi-objective optimization model to determine the sea-containers and trailers to be unloaded on a given day to replenish stock levels such that the detention penalties and the manufacturing warehouse utilization are minimized.
Once a sea-container or trailer is selected to replenish the warehouse inventory levels, its contents (i.e., pallets of parts) must be unloaded by the forklift operator and then processed by workers to update the stock levels and break down the pallets if needed. Finally, the unloaded and processed part is stored in the warehouse bins or shelves. In Chapter 5, we study the problem of determining the optimal team formation such that the total expected time required to unload, process, and store all the parts contained in the sea-containers and trailers selected for unloading on a given day is minimized.
Recommended Citation
Malde, Avnish Kishor, "Optimal Global Supply Chain and Warehouse Planning under Uncertainty" (2022). All Dissertations. 3105.
https://open.clemson.edu/all_dissertations/3105
Author ORCID Identifier
0000-0003-1658-332X