How Dijitalis Saved $1.5 Million with AGV Optimization Simulation in Electronics Manufacturing

Executive Summary

Dijitalis Consulting, a leading simulation and optimization firm, was tasked with optimizing Automated Guided Vehicle (AGV) investments for a global electronics manufacturer . The client was planning significant facility upgrades, including replacing their outdated AGV fleet of 132 vehicles that frequently caused production delays. Using Simio’s powerful simulation capabilities, Dijitalis created a comprehensive digital twin manufacturing model of the 72,000m² facility, including 15 assembly lines, 6 AGV parking areas, and 169 delivery points.

The simulation revealed that only 95 AGVs were required—37 fewer than initially proposed—resulting in capital expenditure savings exceeding $1.5 million. Beyond cost reduction, the Simio model became an invaluable continuous improvement tool, allowing the client to test layout modifications, process changes, and production schedules before implementation. This case study demonstrates how simulation-based decision making can deliver substantial ROI while ensuring operational excellence.

Client Background

Dijitalis Consulting was established in 2006 in Istanbul and has successfully delivered over 250 projects for more than 400 clients across 15 countries. The company specializes in building mathematical models to analyze material flow, identify inefficiencies, and test facility improvements for clients in automotive, manufacturing, logistics, and textile sectors.

The client, a global electronics manufacturer, operates a 34,000m² production facility with 15 assembly lines. Their existing AGV fleet of 132 vehicles was outdated and frequently caused production delays by getting stuck in the path network and failing to deliver materials on time. The company was initiating major investments, including automated warehouses, a new paint shop, increased production capacity, and a new AGV fleet.

Challenge: Beyond Simple Fleet Sizing

The client faced a complex decision regarding their AGV investment. While the primary question was how many AGVs to purchase, the challenge extended far beyond simple fleet sizing:

Vendor Uncertainty

AGV manufacturers typically aim to sell as many vehicles as possible without conducting detailed analysis to determine the optimal number. They rarely develop complex models to prove their recommendations, leaving clients to make decisions based on rough estimates or past experience.

Operational Complexity

The facility’s material handling system was highly complex:

• 15 assembly lines with varying production schedules
• 6 AGV parking locations serving 169 delivery points
• Thousands of different SKUs across 27 material groups
• Frequent changeovers requiring precise material delivery timing

Multiple Interconnected Questions

The client needed answers to numerous interconnected questions:

• Is the production schedule feasible with the new AGV system?
• Will there be stockouts during production?
• When should changeovers occur?
• How should the supply process be designed?
• When should material supply for new orders begin?
• How many AGV loading stations are needed?
• Will there be deadlocks in the path network?
• Which areas will experience high traffic congestion?

Traditional spreadsheet-based calculations couldn’t account for the dynamic interactions between these factors, making simulation the only viable approach for comprehensive optimization.

Solution: Data-Driven Simulation Modeling with Simio

Dijitalis employed a structured approach using Simio simulation software to create a comprehensive digital twin of the facility’s material handling operations:

Data Collection and Validation

The team collected and validated extensive data including:

• Facility layout and dimensions from AutoCAD drawings
• Bill of materials for all products
• Warehouse and material storage locations
• AGV specifications and parking stations
• Production schedules and changeovers

Supply Process Design

After careful analysis, the team designed a push-based supply system rather than a pull system. This decision was based on several factors:

Implementation Practicality: A pull system would require calculating optimal reorder points and quantities for thousands of different SKUs, making it impractical to implement and manage.

Operational Simplicity: The push system required only one parameter—how many minutes before changeover to begin supplying materials for the next product.

Storage Availability: The assembly lines had sufficient storage space to accommodate the push system’s approach.

The process used production plan data, pallet information, and bill of materials to calculate remaining production time and trigger material supply for the next product at the appropriate moment before changeover.

Simulation Model Development in Simio

The Simio simulation software’s object-oriented structure and data-driven capabilities were instrumental in creating an accurate, flexible model:

Data-Driven Approach

The model utilized Simio’s data table functionality to import and manage:

• Production line definitions with coordinates and schedules
• Production plans with order sequences and lot sizes
• AGV parking areas with precise coordinates
• Storage point locations extracted from AutoCAD
• Material routing information for 435 different routings
• Bill of materials for 387 different SKUs

Efficient Model Creation

Simio’s capabilities enabled:

• Automatic creation of 169 supply destination objects at precise coordinates
• Definition of custom library objects that could be modified globally
• Integration of production data through relational database structures
• Use of the Material element to define bill of materials without creating thousands of individual entities

Visualization and Analysis

The model provided powerful visualization capabilities:

• 3D representation of the facility and AGV movements
• Real-time monitoring of AGV utilization and locations
• Heat maps showing traffic density and waiting times
• Detailed tracking of material flow and inventory levels

Model Verification and Improvement

The simulation revealed several design issues that would have caused operational problems:

Deadlock Prevention

The initial design included a bidirectional single lane in a high-traffic area. The simulation demonstrated that this would cause deadlocks as AGVs from opposite directions would block each other. The team recommended changing to single-directional paths in opposite directions, eliminating the deadlock risk.

Traffic Optimization

Heat maps generated from the simulation identified:

• Areas with high traffic volume
• Locations where AGVs spent the most time waiting
• Junction points causing congestion

These insights led to layout modifications that improved traffic flow and prevented bottlenecks.

Scenario Testing and Optimization

Using Simio’s experiment module, the team conducted 35 scenarios testing different combinations of:

• Number of AGVs in each parking area
• Number of loading stations in each area
• Order release time trigger (minutes before changeover)

The primary KPI was production delay, with a target of zero minutes. Secondary KPIs included AGV utilization rates and the number of AGVs available in parking areas as a buffer for maintenance or breakdowns.

Results: $1.5 Million Savings and Operational Excellence

The simulation-based optimization delivered substantial benefits:

Capital Expenditure Reduction

The optimal scenario required only 95 AGVs instead of the initially proposed 132, representing a 28% reduction in fleet size. With an average AGV cost of $50,000, this translated to capital expenditure savings exceeding $1.5 million.

Optimized Resource Allocation

The simulation determined:

• The precise number of AGVs needed in each parking area
• The optimal number of loading stations required in each area
• An order release time of 16 minutes before changeover was sufficient

Balanced Utilization

The optimized scenario achieved average AGV utilization rates of 65-66%, representing an efficient balance between resource availability and operational requirements. The simulation revealed utilization patterns throughout the day, with peaks during morning changeovers and afternoon batch productions.

Zero Production Delays

Most importantly, the optimized configuration ensured zero production delays due to material delivery issues, maintaining production efficiency while minimizing capital investment.

Beyond Initial Optimization: A Continuous Improvement Tool

The value of the Simio model extended far beyond the initial AGV fleet sizing:

Ongoing Decision Support

The simulation model became a continuous improvement tool for:

• Testing layout modifications
• Evaluating process changes
• Validating production schedules
• Assessing the impact of new product introductions

Comprehensive Facility Management

The single simulation project provided multiple tools:

• CapEx and OpEx cost optimization
• Path network deadlock prevention
• Supply management optimization
• Decision support for new investments
• Parameter-based resource optimization

Future-Proofing Operations

As the client’s product portfolio and production requirements evolve, the Simio model continues to provide value by:

• Testing the feasibility of new production schedules
• Evaluating the impact of product changes on material flow
• Identifying potential bottlenecks before they occur
• Supporting data-driven decision making

Conclusion: The Power of Simulation-Based Decision Making

This case study demonstrates the transformative impact of simulation-based decision making on capital investment planning and operational optimization. By replacing rough estimates and vendor recommendations with data-driven analysis, Dijitalis helped their client:

• Save over $1.5 million in unnecessary AGV purchases
• Ensure production efficiency with zero material delivery delays
• Prevent potential bottlenecks and operational issues
• Create a valuable asset for ongoing improvement and decision support

The project showcases how Simio’s powerful simulation capabilities can deliver substantial ROI while providing insights that would be impossible to obtain through traditional analysis methods. For manufacturing and logistics operations facing complex material handling challenges, simulation offers a proven approach to optimize investments and enhance operational performance.