Optimizing Metal Component Supply Chain in Automotive OEM Plant: A Case Study

Thesis Abstract

This study addresses persistent inefficiencies in the metal component supply chain within a high-volume automotive OEM plant, where irregular supplier lead times, quality variability, and suboptimal inventory policies contribute to production delays and increased total cost. The aim is to develop an integrated, data-driven framework for optimizing component sourcing, inventory management, and logistics to improve on-time delivery, reduce working capital, and enhance supplier collaboration. Specific objectives include (1) quantifying the impact of supplier lead-time variability and quality defects on production throughput; (2) calibrating a multi-echelon inventory model that integrates safety stock, lot sizing, and demand forecasting for key metal components; (3) evaluating the effects of supplier collaboration practices and information sharing on supply chain resilience; (4) designing a robust scheduling and transportation plan that minimizes total logistics cost while maintaining service levels; and (5) validating the framework through a pilot implementation and scenario analysis. The study adopts a mixed-methods research design rooted in operations management and supply chain theory, combining quantitative modeling with qualitative insight to ensure practical relevance. The population comprises all metal component suppliers and internal manufacturing units within the target OEM plant, with a sample of 18 suppliers selected via stratified random sampling and 3 internal manufacturing departments chosen through purposive sampling to capture process diversity. Data collection instruments include ERP-generated operational data (12- to 24-month history of order quantities, lead times, defect rates, and production stoppages), supplier performance records, on-site process observations, and semi-structured interviews with 14 procurement and logistics personnel and 6 supplier relationship managers. Instrument validity is established through content validity with domain experts and triangulation across data sources; reliability is assessed via test-retest procedures for the interview protocol and Cronbach’s alpha for survey-like instruments. Analytical techniques comprise a combination of descriptive statistics, time-series analysis, and regression modeling to quantify relationships between lead-time variability, quality defects, and production performance. A multi-echelon inventory model (including base-stock and reorder-point policies) is formulated and solved using nonlinear programming to determine optimal safety stock and ordering policies under stochastic demand. Scenario analysis and Monte Carlo simulation are employed to assess resilience under disruption and to compare the baseline against proposed policy changes. A cooperative game-theoretic framework is used to model supplier collaboration incentives and information-sharing configurations, and structural equation modeling (SEM) tests the causal links among collaboration, information sharing, and performance outcomes. Thematic analysis of interview data identifies organizational enablers and barriers to implementing the proposed framework, guided by the resource-based view and the theory of constraints. Expected findings indicate that (a) reducing lead-time variability and improving first-pass quality can significantly decrease production stoppages and buffer stock requirements; (b) the integrated inventory policy yields lower total cost of ownership for critical metal components while maintaining or improving service levels; (c) explicit supplier collaboration mechanisms and information-sharing protocols enhance supply chain responsiveness and reduce bullwhip effects; (d) an optimized logistics schedule decreases inbound transportation costs by a measurable margin without compromising throughput. The study contributes to knowledge by operationalizing an end-to-end optimization framework for automotive metal components, integrating quantitative supply chain modeling with organizational and relational factors, and demonstrating the practical viability of coordinated procurement, production planning, and logistics in a real-world setting. The main conclusion posits that aligned supplier collaboration, data-driven inventory optimization, and disciplined logistics scheduling jointly yield substantial performance gains in automotive metal component supply chains. Recommendations include institutionalizing a formal supplier integration program, investing in real-time data analytics and visibility platforms, and adopting the proposed multi-echelon model as a standard decision-support tool, with further research suggested to generalize the framework to other material families and to explore digital twin implementations for continuous improvement.