Case Study: BMTA Fleet Management - Performance Test

Background

The Bangkok Mass Transit Authority (BMTA) operates a fleet of over 5,000 vehicles, serving millions of passengers across Bangkok and its metropolitan area. To ensure efficient fleet operations, BMTA implemented a Fleet Management System to monitor vehicle usage, optimize routes, and streamline maintenance scheduling.

However, as BMTA's operations expanded, system performance issues began to emerge under peak loads. Delays in fleet data updates, system crashes, and slow response times negatively impacted operational efficiency.

Challenges

1. High Data Traffic:

Data from 5,000+ buses sent GPS updates every second, creating heavy incoming traffic.

2. Real-Time Data Processing

Processing and visualizing GPS data in real time for operators, requiring seamless database and application performance.

3. Large API Payloads

Each API call transmitted payloads between 200KB and 300KB, creating bottlenecks under heavy load.

Objectives

checkDetermine the system’s maximum capacity and its performance under normal operating conditions.

checkIdentify and resolve bottlenecks related to high traffic, real-time data processing, and large payloads.

checkProvide actionable recommendations to improve response times, data processing, and scalability.

Approach

1. Requirement Analysis

Identified the system’s breaking point by simulating increasing traffic and payloads. Evaluate the system performance under normal usage.

2. Performance Test Strategy

Stress Test- Simulated extreme traffic conditions to identify the system’s breaking points.- Gradually increased the load to evaluate system behavior under duress.

Load Test- Tested the system under normal and peak loads to measure its ability to maintain stability and performance.- Focused on key metrics such as response time, throughput, and error rates.

3. Load Testing Execution

Used k6 performance testing tool, to simulate high data traffic and large payloads.

4. Data Collection, Analysis, and Reporting

- Monitored performance metrics.- Analyzed root causes of system failures and identified optimization opportunities.- Delivered a detailed report with actionable recommendations.

Findings

1. Concurrency Issues

2. High Data Traffic

3. Large API Payloads

Solutions

1. Using Pagination to Manage High Data Traffic

2. Optimizing API Payloads

Results

checkEnhanced Concurrency: System capacity improved from 20 concurrent users to 100 concurrent users without crashes.

checkReduced API Response Time: Response times dropped from 5 seconds to under 2 second, even with large traffic volumes.

checkSmaller API Payloads: Payload sizes reduced by 30%, improving network and system performance.

Case Studies

Discover how Asheon has transformed businesses with innovative, scalable, and tailored solutions. Our case studies showcase real-world examples of successful projects, highlighting challenges faced, solutions implemented, and measurable results delivered.

    Read Our Case Studies

      Let's Bring

      Your Ideas to Life

      Whether you’re starting fresh or refining your vision, we’re here to bring it to life. Let’s collaborate and build your perfect solution together!