Optimizing Supplier Selection and Inventory Replenishment Using Hybrid Analytical Models in a Global Manufacturing Network

Project Abstract

In global manufacturing networks, the dual challenges of supplier selection and inventory replenishment demand integrated decision-making that aligns procurement strategy with dynamic demand, supplier risk, and multi-echelon logistics. This study develops a novel hybrid analytical framework that combines data-driven forecasting, optimization under uncertainty, and robust supplier evaluation to minimize total cost of ownership while maintaining service levels across distributed plants. We begin by constructing a comprehensive data pipeline that integrates internal transaction data, supplier performance metrics, macroeconomic indicators, and real-time market signals. Time-series models, including advanced machine learning regressors and ensemble methods, are deployed to forecast demand at multiple horizons and to capture seasonal, promotional, and structural shifts. These forecasts feed a multi-objective optimization model that coordinates supplier selection, lot sizing, order quantities, and safety stock across the network, with objectives encompassing cost, sustainability, risk exposure, and supplier diversity. A core contribution is the integration of hybrid analytical models (i) a probabilistic demand forecasting module that quantifies forecast uncertainty and propagates it through the replenishment decisions; (ii) a robust optimization layer that accommodates supply disruptions, lead-time variability, and capacity constraints using scenario-based and distributionally robust techniques; (iii) a supplier evaluation mechanism that blends qualitative risk indicators with quantitative performance data into a scoring framework, guiding procurement strategy under evolving market conditions; and (iv) a feedback loop that leverages reinforcement learning to adapt policy parameters based on realized service levels and costs. The framework supports decoupled and integrated procurement strategies, enabling negotiable strategies for key suppliers while maintaining operational resilience for non-critical sources. A case study across a global electronics manufacturer demonstrates the model’s ability to reduce total operating costs by X% and improve service level compliance by Y% under various disruption scenarios, including supplier bankruptcy risk, port congestions, and demand surges. Sensitivity analyses reveal the relative impact of forecast accuracy, lead-time variability, and supplier risk on overall performance, informing resource allocation for data quality improvements and supplier development programs. The study also assesses environmental and social governance implications by incorporating carbon footprint and supplier diversity into the objective function, illustrating how hybrid models can reconcile cost minimization with sustainable procurement goals. The proposed framework is implemented using open-source tools and scalable optimization solvers, ensuring practical transferability to organizations with diverse product portfolios and global supplier bases. The findings provide actionable guidelines for procurement managers on selecting suppliers, configuring inventory policies, and designing resilient supply chains that sustain competitiveness in volatile global markets.

Project Overview

What This Project Is About

---

A plain-language overview of the topic and what the project investigates.

---

---

The Problem It Addresses

---

What problem or gap this project tackles and why it matters to the field or society.

---

---

Objectives of the Project

---

1. Identify criteria for choosing suppliers in a global network.
2. Develop replenishment rules to reduce stockouts and excess inventory.
3. Integrate simple analytical tools to support decision making across procurement and operations.
4. Assess how supplier risk and demand variability affect supply chain performance.
5. Provide practical guidelines for implementing hybrid models in real settings.

---

---

What You Will Do Step by Step

---

1. Review basic concepts in purchasing, inventory, and supplier management.
2. Collect or simulate data on supplier performance, lead times, and demand.
3. Apply simple analytical methods to rank suppliers and set reorder points.
4. Test replenishment strategies under different demand and disruption scenarios.
5. Evaluate outcomes using easy-to-understand metrics like cost, service level, and total cost of ownership.
6. Document decisions and propose practical implementation steps for a real setting.

---

---

Expected Outcome

Clear guidance on selecting suppliers and timing orders that balance cost and service, with a framework anyone can adapt in a real company.