Case Studies > Transportation Success Story

Transportation Success Story

Company Size

1,000+

Region

America

Country

United States

Product

Aspen Mtell

Tech Stack

Machine Learning

Implementation Scale

Enterprise-wide Deployment

Impact Metrics

Cost Savings

Technology Category

Analytics & Modeling - Machine Learning

Applicable Industries

Railway & Metro

Applicable Functions

Logistics & Transportation

Use Cases

Predictive Maintenance

Services

Data Science Services

About The Customer

The customer is a major U.S. transportation company responsible for delivering goods on time, safely, and reliably. The company operates a large fleet of locomotives and is committed to ensuring the highest levels of operational efficiency and reliability. However, the company was facing significant challenges due to undetected catastrophic failures of locomotives. These failures were not only costly in terms of repairs and additional operational costs but also resulted in fines and disruptions in the delivery of customer goods. The company needed a solution that could help it detect these failures in time and prevent them from occurring.

The Challenge

A major U.S. transportation company was facing significant losses due to undetected catastrophic failures of locomotives. These line-of-road (LoR) engine failures were costing the company over a million dollars each in repairs, additional operational costs, and fines. The company's existing reliability techniques were not sufficient to detect these failures in time, leading to disruptions in the delivery of customer goods and impacting the company's reputation for safety and reliability.

The Solution

The company partnered with AspenTech and implemented their product, Aspen Mtell. Aspen Mtell is a machine learning-based solution that was used to examine laboratory analysis data from engine lube oil samples. Through multi-variate machine learning analysis of archived samples from the 30 “bad” actors, Aspen Mtell insight discovered normal behavioral patterns and exact failure patterns. These insights were then transferred to Agents executing on as many as 600 locomotives. The solution provided the company with the ability to detect potential failures in advance, allowing them to take preventive action and avoid costly repairs and operational disruptions.

Operational Impact

Within four months of implementing the Aspen Mtell solution, the company was able to prevent 10 major locomotive failures. This was achieved through the foresight alerts provided by the Aspen Mtell Agents, which included prescribed action to prevent the failures.
In one instance, even when an engine successfully passed a low-pressure test, the Aspen Mtell application engineer noticed a failure signature and alerted management. A subsequent high-pressure test showed leaks in eight places, preventing an expensive, looming LoR failure.
After scaling the Aspen Mtell solution to more than 4,000 locomotives, the company was able to detect failures on 96 engines, resulting in additional major savings.

Quantitative Benefit