top of page

AI Case Study

IBM launches Equipment Maintenance Assistant to predict equipment maintenance requirements and aid technicians in diagnosis and repair using machine learning

IBM has launched a machine learning and natural language processing capable system called Equipment Maintenance Assistant. It purportedly is able to ingest and learn from a variety of maintenance repair documents while taking input from machinery sensors to help field technicians diagnose and repair equipment issues.

Industry

Technology

Software And It Services

Project Overview

"IBM Maintenance Equipment Assistant augments your asset maintenance program with machine learning techniques and AI tools. This powerful combination provides asset-intensive industries with capabilities to optimize asset repairs based on prescriptive guidance. AI methods using IBM Watson technology are applied to structured and unstructured data associated with repairs, maintenance, procedures and techniques. This offers enhanced insights and recommend optimum repair methods and procedures. It enables equipment manufacturers to detect failure patterns, ensure optimal first-time fixes, and extend the life of critical assets."

"Our solution builds its knowledge base using your structured and unstructured data associated with repairs, work order history, technical documents, specification diagrams, industry blogs, and more. Based on that data, it identifies early warning signs and identifies the optimal repair or service procedure for your technicians to execute. It also continuously learns from the technicians’ interactions and uses this feedback to improve future recommendations."

Reported Results

IBM claims its Equipment Maintenance Assistant

"• Improves first time fix rates – improves efficiency of repair by up to 10 percent through prescription of the right repair procedures based on your data and AI
• Extends asset life – extend asset life with step-by-step repair guidance and enable compliance with current standards, regulations and best practices
• Reduces time to repair – complete repairs faster through more efficient troubleshooting and prescriptive repair guidance
• Increases time between failures – optimize repair procedures to extend equipment operating time and decrease unplanned downtime"

Technology

"Technician repair guidance is made faster and easier with a conversation based user interface including speech to text and text to speech. This means technicians can engage with guidance naturally in a method that supports their demanding roles.
Based on IBM Watson AI technology including the ability to continually learn over time. This means technician feedback is continually incorporated which sharpens repair guidance and improves outcomes."

Function

Strategy

Planning

Background

"A large portion of the workforce is retiring – which will have major implications on the knowledge base in many organizations. In particular, many experienced technicians are exiting without proper knowledge transfer mechanisms in place. Even worse, much of the knowledge lost is intrinsic and very difficult to regain.
The number and complexity of critical assets is increasing. These assets also generate more data and are more complex to repair. Being able to capture, analyze and use this data is critical to guide less experienced technicians and to reduce maintenance costs."

Benefits

Data

"Can ingest both structured and unstructured data from a variety of sources to provide optimal repair guidance.. Types of inputs include repair manuals, work orders, written technician comments, root-cause failure analysis, and available regulatory standards documentation."

bottom of page