Benefits of unlocking AI-powered maintenance for manufacturers

The rise of AI-powered CMMS presents an urgent opportunity for organizations to adopt fully digital solutions, but it needs reliable data capture of yesterday's actions to fully realize its potential.

By Nick Haase June 26, 2024
Courtesy: MaintainX

 

Learning Objectives

  • Understand how outdated maintenance practices cost manufacturers trillions due to unplanned downtime.
  • Learn how AI-powered CMMS can improve maintenance but relies on accurate data from past actions.
  • Understand why companies must adopt digital maintenance solutions to capture institutional knowledge and prepare for future AI advancements.

AI-powered maintenance insights

  • Outdated maintenance practices cost manufacturers significantly, underscoring the need for AI-powered predictive maintenance to address issues proactively before they escalate, improving efficiency and reducing costs.
  • Transitioning to mobile-first CMMS and AI solutions is vital for capturing comprehensive maintenance data, enabling better predictive maintenance, reducing downtime and adapting to workforce changes.

According to a recent ABB report, outdated maintenance and reliability practices are a trillion-dollar problem for manufacturers, with unplanned downtime alone costing a staggering $125,000 per hour. This financial burden underscores the critical need for AI-powered predictive maintenance solutions that can proactively address issues before they escalate. However, most maintenance teams face significant data-related challenges in implementing these advanced tools.

Root causes of maintenance data challenges

Traditional maintenance practices, ranging from pen and paper to emails, spreadsheets and whiteboards hinder efficiency and increase risk. Consider the Boeing incident earlier this year. The maintenance records were irretrievable, making it difficult to ascertain if the issue was preventable and conduct a root-cause analysis (RCA).

While many manufacturers have deployed legacy computerized maintenance management systems (CMMS), shop floor technicians struggle to adopt these systems as they often work offline and away from a desktop computer. Their inputs are captured using pen and paper, and are often added manually after tasks are completed, causing delays, data gaps and inaccuracies. This drives the perception CMMS solutions are cumbersome and not worth it.

However, some of the industrial software systems have neglected the needs of mobile, deskless workers, whose input is crucial for success. This results in systems defaulting backward into paper or manual-dependent processes with, at best, marginal computer-based guidance. This disconnect results in a lack of valuable maintenance and enterprise asset management (EAM) data, which rely on two-way data flow to optimize recommendations.

For example, such a system might have a record of all downtime events but not include machine conditions just before or in the weeks leading up to the downtime event, preventing condition-based maintenance analysis. The system might record all maintenance procedures for a specific repair but not include SKUs utilized and the resulting impact on inventory. There might be a record of the procedures for planned maintenance, but a lack of an audit trail of the instances and ways those procedures were adjusted and why.

The rise of AI-powered CMMS presents an urgent opportunity for organizations to adopt fully digital solutions and capture as much data as possible. AI-enabled, end-to-end CMMS promises to improve predictive maintenance today, but the accuracy of its predictions depend on reliable data capture of yesterday’s actions to power AI training. If organizations applying the manual-based processes mentioned above don’t recalibrate, they won’t have the data foundation required to act on the AI innovations.

Today, even organizations behind on their digital strategy can leapfrog to mobile-first CMMS, turning cost centers into performance drivers. Many of those first to deploy CMMS built customized applications with in-house teams building them off of existing enterprise software solutions, such as enterprise resource planning (ERP) modules or SharePoint. These teams might have extensive maintenance data and metadata recorded, but in static formats where utilization can require engineering expertise or deployment of a business intelligence solution to help visualize and activate the stored information. The in-house team also must continually update, patch and maintain the system.

Companies adopting CMMS today benefit from systems with built-in visualization, collaboration tools, pre-built integrations and access to expert support. There also is an opportunity to capture invaluable institutional knowledge before longtime team members head for retirement while also enriching their organization’s training data for the future. A user-focused, technology-driven maintenance approach in this environment becomes more than an opportunity; it’s a corporate and fiduciary duty.

Organizations behind on their digital strategy can leapfrog to mobile-first computerized maintenance management systems (CMMS), turning cost centers into performance drivers.

Organizations behind on their digital strategy can leapfrog to mobile-first computerized maintenance management systems (CMMS), turning cost centers into performance drivers. Courtesy: MaintainX

CMMS isn’t new. What is?

For those using CMMS, ask a simple question: Do you trust the quality of the data in your CMMS today? A negative answer likely indicates the field team lacks the tools or motivation to accurately capture data.

Frontline technicians are inundated with too many disconnected and unprioritized alerts, alarms, and work orders across multiple apps and documentation sources, with complicated interfaces and long load times. This leaves them with a simple choice: Do the work and move on to the next task, or spend 30 minutes figuring out how to record their work most accurately, thereby reducing wrench time.

This may sound hyperbolic, but it’s a reality on shop floors around the world every day. The friction leads to a situation where most operational data remains unlogged, hindering its use for future organizational improvements.

Digital maintenance technology, especially AI-powered tools, offers the most value when they’re designed with technicians in mind. If the software is user-friendly, helps them do their jobs better and fits naturally into their existing workflows, users are more likely to adopt it. Companies create a valuable foundation for AI models by consistently capturing data through this software. These models can then better learn ideal and non-ideal operating conditions and maintenance procedures and analyze the data to uncover risks that need to be addressed, as well as opportunities for streamlining operations and making improvements.

To build a solid foundation for AI training data, companies must focus on capturing comprehensive daily work execution and asset data. This includes logging every maintenance task, tracking equipment performance, inventory, time and procedures utilized in repair. They also need to record contextual environmental conditions for a site’s most critical assets and processes.

When teams use a CMMS, every job, interaction and other maintenance touchpoint produces valuable data that can provide the groundwork to better understand maintenance cycles. While logging every work order might seem tedious, investing in streamlined data practices can yield significant savings in downtime, resource planning, anomaly detection, safety and more.

For example, building materials producer Titan America deployed CMMS to help establish a reliability-centered maintenance (RCM) program in its Florida plant and drive measurable O&M savings. A key requirement was user experience. With an intuitive CMMS, its technicians could easily follow digitized inspection checklists from their mobile devices, enter work order data in real time, and escalate issues to supervisors via instant messaging within seconds. The solution brought new access and ease, which helped drive adoption and increase data capture. In less than a year, not only did the team eliminate over 30% of unplanned downtime, but they’re also building a data foundation for AI-enabled predictive maintenance.

Businesses that don’t capture data everyday results in a loss of potential efficiency and the opportunity to improve future AI and related use cases. Building AI models that understand a business’s, plant’s and team’s maintenance needs is vital. The best digital maintenance solutions today enable the better solutions of tomorrow.

Capturing institutional knowledge quickly

AI is entering just as the industry faces massive workforce changes within maintenance teams. The U.S. Bureau of Labor Statistics projects 3.1 million jobs in maintenance and repair occupations will be added between 2021 and 2031 due to retirements and new growth requirements for maintenance. More urgently, the 2023 National Association of Manufacturers’ (NAM) report, “The Skills Gap and Workforce Needs Survey,” found 28% of manufacturers expect more than 25% of their workforce to retire within the next five years.

New generations integrating into the workforce come with a technological background and expectation they will be digitally enabled. There is little time to ensure they are personally trained across each procedure or piece of equipment by the folks about to retire.

In a recent Harris Poll survey of working Baby Boomers, 57% said they have shared half or less of the knowledge needed to perform their job responsibilities with those who will assume them. Senior technicians must impart decades of experience while doing their day job – to train both new employees and the AI solutions that will enable them and future teams.

A CMMS is a knowledge hub that enables all stakeholders to input and access critical data. It helps solve urgent near-term issues around onboarding for frontline professionals and knowledge capture before experienced technicians leave while creating a more trusted and complete data lake that powers AI training.

A comprehensive CMMS platform must excel at making work easier today and having its features offer scalability to meet business needs as they grow and change.

Future winners are being determined now

The challenges posed by legacy maintenance systems, coupled with the looming workforce changes and the rising expectations of stakeholders, underscore the urgency for a digital transformation. However, the potential of AI at this juncture means CMMS adoption is not just a solution to these immediate challenges; it’s a strategic imperative for long-term growth and viability. The future of maintenance is being shaped today. Companies that embrace digital transformation and AI-powered solutions will not only survive but thrive in the evolving industrial landscape.


Author Bio: Nick Haase is a MaintainX co-founder and he has built several companies over his career and actively advises early-stage companies on marketing and sales strategies.