Predictive Maintenance vs. Preventive Maintenance: Understanding the Difference
Proactive rather than reactive maintenance is quickly becoming the norm for enlightened facility...
If you’re reading this article, chances are you’re considering implementing a predictive maintenance program. You have probably read about what predictive maintenance is, and you’ve most likely explored the benefits of predictive maintenance. Now, you’re likely considering what steps you’ll need to take to implement such a program. The first step, however, is to understand the challenges of implementing an effective predictive maintenance strategy so that you can devise a plan to overcome them.
Before we get started if you need a refresher or haven’t explored the topics mentioned above, read our Ultimate Guide to Predictive Maintenance to catch up.
A predictive maintenance strategy is only as good as the data that it uses. Many building maintenance plans that tout a predictive maintenance strategy still rely on human input for the necessary data that informs predictive maintenance. With human input comes human error. This can significantly impact the success of your predictive maintenance program.
Facility management’s slow adoption of technology-driven data collection is not without cause. The root of the problem is a web of data inconsistencies in the built environment. Getting the HVAC system to connect with the occupancy sensors and the access control system and the lighting system isn’t as easy as just hooking them all up to the same software. Each system tends to operate within its own ecosystem of languages, databases, user interfaces, and reporting tools. To achieve the benefits of predictive maintenance each of these ecosystems needs to be standardized.
In recent years, the solutions to this problem have progressed. However, data and relationship modeling remain highly inconsistent. The underlying technology and modeling solution must solve the myriad of naming protocols, data standards, ontologies, and relationship models that exist across the different systems and equipment within a building. It must do all of this with an eye to the future, anticipating that new technologies emerge exponentially and will need to be quickly and easily added to the ecosystem to continuously improve the P-F curve model’s accuracy.
Solving the Data Challenge
While the data challenge is a significant barrier to implementing a predictive maintenance program, there are technology solutions that make it easy to solve. Implementing an intelligent building management platform (IBMP) that leverages an Independent Data Layer (IDL) to simplify the web of data inconsistencies is an easy way of overcoming this challenge.
Many maintenance engineering leaders make the critical mistake of adopting a predictive maintenance program and turning it on once everything is hooked up.
Unfortunately, if your predictive maintenance data is accurate and reliable, this usually results in a tsunami of alarms and work orders that quickly engulfs maintenance teams. As maintenance teams scramble to address the sometimes thousands of new alarms, critical failures often get lost in the noise.
This can have catastrophic effects on the success of your program not only because of the implications of missing critical alarms but can also sour maintenance teams’ appetites for the program altogether.
Solving the Processes and Procedures Challenge
You should establish a set of priorities, and even better, implement them into a software solution that automatically prioritizes alarms based on that set of priorities. This can help your team systematically manage the influx of problems that a predictive maintenance program can uncover.
New software often requires new skill sets, new training, and new responsibilities. Particularly, predictive maintenance relies on the interpretation of data to uncover anomalies in equipment performance, external factors, and trends that signify future failures. This is typically a different skill set than what your maintenance engineering team most likely possesses.
Aside from the learning curve associated with new software, new technologies require changes to daily activities. This can result in your team’s reluctance to assimilate into the new program and can hinder its long-term success.
Solving the Adoption Challenge
Several solutions – such as onPoint – lessen the need for significant reskilling. These solutions provide insights developed by subject matter experts in both data science and building systems that help guide building maintenance teams to address issues with minimal need to obtain new knowledge. Therefore, it’s important to choose a provider with experience beyond software development. You should also seek a software provider with experience in building systems and controls. That way, the insights produced from your software are sensible root-cause findings of operational issues that your team understands.
Managed services – such as Buildings IOT’s Digital Services – support these technologies and insights by augmenting facility personnel with subject matter expertise to monitor, analyze and enact predictive maintenance solutions across portfolios. These services collect multiple sources of equipment data and advise on equipment utilization and operation to minimize disruption, costs, and complaints.
Now that you’ve familiarized yourself with the challenges you’re likely to face in implementing a predictive maintenance program, the next step is to review the necessary steps for implementing the program itself. If you’re ready to take this step, our team at Buildings IOT is an excellent resource. Schedule an introductory call with one of our Smart Building Experts to learn more.
Based in Dayton, Ohio, Brian Cline is the digital service manager and a Master Systems Integrator at Buildings IOT. Brian holds over 20 years of experience in project management and technical solutions implementation of building automation systems.