Optimization of a Lean Manufacturing System for Small-Scale Automobile Components using Digital Twin Simulation

Project Abstract

This study presents a comprehensive framework for optimizing a lean manufacturing system tailored to small-scale automobile component production by leveraging digital twin simulation. The research integrates lean principles—value stream mapping, pull production, set-up time reduction, and continuous flow—with advanced digital twin technologies to model, simulate, and optimize end-to-end operational performance. A modular digital twin of the manufacturing system is developed, capturing real-time data from shop-floor processes, material handling, machine performance, and quality outcomes, enabling rapid scenario analysis and decision support. The methodology begins with a thorough current-state assessment to identify waste, bottlenecks, and variability sources within the production line, followed by a future-state design that aligns with lean objectives and resource constraints typical of small-scale enterprises. Key components of the digital twin include discrete-event simulation for process flow, agent-based models for human-robot interactions, and physics-based models for machining and assembly operations, integrated within a cloud-enabled platform for scalability and collaboration. Data-driven calibration ensures accurate representation of cycle times, downtime, yield, and setup durations, while stochastic optimization techniques explore trade-offs between production lead times, inventory levels, and service levels. The optimization process targets reductions in changeover times, batch sizes, and work-in-process, while improving overall equipment effectiveness and line throughput under fluctuating demand. A novel hybrid approach combines lean metrics (OEE, Takt time compliance, SMED) with digital twin-driven optimization to generate actionable improvements such as line reconfiguration, kanban sizing, buffer management, preventive maintenance scheduling, and automated quality feedback loops. The study also incorporates risk assessment and resilience planning to mitigate disruptions from supply variability and machine failures, ensuring robust performance of the lean system. Validation is performed through a multi-stage process, including face validation with industry experts, historical data back-testing, and forward-looking simulations under multiple demand and disruption scenarios. The results demonstrate significant performance gains accelerated product mix changeovers, reduced inventory carrying costs, improved on-time delivery, and higher first-pass yield, while maintaining safety and ergonomic standards. Sensitivity analyses identify critical drivers of system performance and quantify the benefits of digital twin-enabled visibility for continuous improvement. The research contributes a replicable methodology for small-scale automotive manufacturers seeking lean agility through digital twin technologies, including a structured data architecture, model libraries, and an implementation roadmap that addresses organizational readiness, data governance, and change management. The study also discusses limitations related to data quality, model complexity, and the integration of legacy systems, offering practical guidelines for incremental deployment. In summary, the fusion of lean manufacturing with digital twin simulation provides a powerful toolkit for optimizing operations, reducing waste, and enhancing competitiveness in the niche of small-batch automobile component production.

Project Overview

What This Project Is About

---

A practical study of how lean manufacturing concepts can be applied to create a more efficient, waste-reducing production flow for small runs of car components. It uses a digital twin—an accurate computer model of the manufacturing process—to simulate, test, and improve the system without changing real equipment.

---

---

The Problem It Addresses

---

Small-scale automotive suppliers often struggle with high setup times, excess inventory, and inconsistent production due to variability in demand. Existing methods may be too costly or risky to test on the shop floor. The project aims to bridge this gap by using digital simulation to identify improvements before implementing them.

---

---

Objectives of the Project

---

1. Identify key sources of waste in a small-scale auto component line.
2. Develop a digital twin of the current production process.
3. Test lean improvements in the virtual model (e.g., pull systems, Kanban, layout changes).
4. Recommend practical changes that reduce lead time and inventory.
5. Demonstrate potential cost savings and productivity gains.

---

---

What You Will Do Step by Step

---

1) Map the existing process and collect data on cycle times, setup times, and defect rates. 2) Build a digital twin model of the line. 3) Validate the model against real data. 4) Run simulations with lean changes. 5) Analyze results and compare scenarios. 6) Propose a prioritized implementation plan. 7) Prepare a final report and presentation.

---

---

Expected Outcome

Clear recommendations for lean improvements supported by simulation results, including a quantified plan for reducing lead times, lowering work-in-process, and improving reliability in small-scale production of automobile components.