Optimizing Sustainable Production Scheduling and Inventory Management in a Mixed-Model Assembly Line Using Digital Twin and Real-Time Data Analytics

Project Abstract

Optimizing Sustainable Production Scheduling and Inventory Management in a Mixed-Model Assembly Line Using Digital Twin and Real-Time Data Analytics integrates advanced modeling, simulation, and analytics to achieve resilient and efficient manufacturing operations. This study addresses the dual challenges of variably configured production lines and fluctuating demand by developing a digital twin framework that mirrors the physical factory in real time, incorporating sensor feeds, machine health indicators, and material flow data. The core objective is to minimize total production and inventory costs while simultaneously reducing energy consumption, lead times, and work-in-progress (WIP) through intelligent scheduling, dynamic lot-sizing, and adaptive replenishment policies. A hybrid optimization approach is proposed, combining mixed-integer linear programming (MILP) for strategic planning with metaheuristic algorithms such as genetic algorithms and particle swarm optimization for tactical and operational decisions. The model accommodates the complexity of mixed-model assembly lines, including sequence-dependent setup times, part obsolescence risk, finite production capacities, stochastic demand, and configurable routing. Real-time data streams from the digital twin enable rolling horizon re-planning, ensuring responsiveness to perturbations such as machine breakdowns, supply delays, and quality excursions. The integration of real-time analytics with digital twin simulations supports proactive maintenance scheduling and condition-based monitoring, thereby increasing equipment availability and reducing unplanned downtime. The research advances a comprehensive data architecture that harmonizes edge devices, cloud storage, and advanced analytics platforms to support scalable decision-making. It introduces a multi-objective optimization framework that simultaneously optimizes operational efficiency, sustainability metrics (energy intensity and waste reduction), and service level performance (on-time delivery and fill rate). Sustainability is embedded through lifecycle-aware inventory policies, dynamic lot sizing aligned with demand forecasts, and energy-aware routing of production orders to minimize peak electricity consumption and carbon footprint. To validate the proposed framework, a high-fidelity industrial-case study is conducted on a representative mixed-model assembly line with heterogeneous product families, varying routes, and shared resources. Scenarios simulate demand volatility, machine reliabilities, and supply disruptions to assess robustness, adaptability, and the trade-offs among cost, service level, and environmental impact. Performance metrics include total cost, lead time, WIP, on-time delivery, machine utilization, energy consumption, and carbon emissions, benchmarked against traditional push systems and static scheduling approaches. The findings demonstrate that the digital twin-enabled, real-time data analytics approach yields substantial improvements reduced total production cost by a significant margin, shorter lead times, lower WIP and throughput times, improved on-time delivery, and measurable decreases in energy usage and emissions. Sensitivity analyses reveal the resilience of the solution to demand shocks and equipment failures, while ablation studies isolate the contributions of digital twin fidelity, forecasting accuracy, and optimization hybridization. The work provides actionable guidelines for practitioners on implementing digital twin-driven scheduling and inventory management in mixed-model environments, along with a roadmap for extending the framework to cross-site and supply-chain wide applications.

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 how to schedule production on a mixed-model line efficiently while reducing waste and downtime. 2. Explore how real-time data can improve inventory decisions and reduce stockouts or excess inventory. 3. Demonstrate how digital representations of the factory (digital twin) help plan and test schedules before implementation. 4. Propose practical guidelines for managers to adopt sustainable production practices.

---

---

---

What You Will Do Step by Step

---

1. Review basic concepts of production scheduling, inventory management, digital twins, and real-time data. 2. Collect or simulate data from a mixed-model assembly line (production orders, part flows, lead times). 3. Build a simple digital twin model of the line to visualize flows and test scenarios. 4. Create a few schedule and inventory policies and compare their performance using the data. 5. Analyze results to see trade-offs between cost, speed, and sustainability. 6. Draft practical recommendations for implementation in a real factory. 7. Prepare a concise project report and a short presentation.

---

---

---

Expected Outcome

A clear set of scheduling and inventory guidelines that use digital twin and real-time data ideas to improve efficiency, reduce waste, and support sustainable manufacturing practices. The project should deliver a simple model, results comparison, and actionable steps for industry adoption.