How do you measure and communicate the value and impact of predictive maintenance?

As a lecturer in predictive maintenance implementation and an advisor to over 70 companies via Power-MI, I've pinpointed the following key objectives and metrics for economic success in PdM: - Downtime Reduction: Aim for a X% decrease in unplanned downtime. - Equipment Availability: Target a Y% boost in machine readiness. - Maintenance Cost Savings: Cut maintenance expenses by Z%, optimizing the budget. These targets are not just theoretical; they're grounded in the strategic financial goals of the firms I mentor. Continuous review and strategy tweaking are essential to hit and exceed these benchmarks.

Your goal in business is always to end up with money in the pocket at the end of the day and the more the merrier. If your downtime and equipment replacement costs are high predictive maintenance is likely something you ll want to invest in especially in today's world of long lead times for replacement parts or equipment....... time is money. Easiest way to provide an example of predictive maintenance we all understand easily is changing the oil in your vehicle regularly to prevent blowing up your motor . We rather drain our engine oil then our bank account . But to go beyond this we can annalise our oil now to see if we haven't changed it soon enough or too soon to get the best out of our oil changes .Changing too soon we waste on oil etc

I completely agree that defining goals and metrics is crucial before implementing PdM. Setting specific outcomes and benefits, along with measuring them, can help achieve success. It is essential to align these goals and metrics with the organization's strategic objectives and KPIs. Establishing baselines and benchmarks for comparison can also help track progress and ensure the success of PdM implementation.

As someone who works in Predictive Maintenance, setting goals and metrics to measure its value is not just another topic, it's the cornerstone of any successful PdM endeavor. A smart way of defining goals is emphasizing how predictive maintenance aligns with broader business goals such as: - improving customer satisfaction - increasing competitiveness - achieving sustainability targets. When it comes to metrics, my experience helped me identify the following as the most relevant: - Equipment uptime - Maintenance costs - Mean time between failures (MTBF) - Return on investment (ROI). Matching the right objectives with the right metrics will position a predictive maintenance project as a strategic enabler of organizational objectives.

Measuring the impact of predictive maintenance hinges on calculating avoided costs, which is complex. A simplified approach involves evaluating overall maintenance cost reductions for large projects. More accurately, for each issue predicted and prevented, calculate the ROI, factoring in savings on spare parts, reduced production losses, and labor hours. This dual approach offers a tangible measure of predictive maintenance's impact, despite its inherently predictive nature.

Understanding the difference between OT (sensor data) and management data is crucial in predictive maintenance. From equipment monitoring, OT data feeds into machine learning for predictive alerts. Management data, encompassing maintenance logs and failure records, is foundational for strategic PdM. This is where we derive KPIs like MTBF and TTR, which are essential for assessing the effectiveness of our maintenance strategies. The challenge lies in accumulating sufficient historical data—a necessity for meaningful analysis and a hurdle in fostering a culture that values long-term benefits over immediate results. Predictive maintenance demands a forward-looking approach: recording now for future insights.

This may require developing (or buying!) new skills. Not only having the understanding of the data, but where it came from, how often it is collected, and what the implications of the format may be. Then, using data science/analysis tools to gain insights. These may be missing skills in your organization.

1) Learn and ask how others did it previously. 2) Be aware of your real data collection and analysis capabilities. Define your procedures accordingly. Sometimes "old school" methods could be enough. 3) Collect data based on your Preventive Maintenance intervals. Try to use only company CMMS to store the data. 4) Compare the changes between observations and compare them to the deterioration rate estimation... recalculate and reset the threshold and collection point as many times as necessary. 5) Standardize the process.

Calculating PdM ROI covers two key areas: projecting for new investments and evaluating current systems. For new investments, we estimate initial costs (hardware, software, training, calibration) versus expected benefits (higher production, lower downtime, reduced maintenance expenses). For existing setups, we gauge savings through scenario comparison (timely action on potential failures), reduced unplanned downtime, extended intervals between planned downtimes, and possible production boosts from effective PdM.

This is a very difficult number to arrive at. The hardest part is knowing where you are starting. You will need to know what your current downtime, availability, etc. are at before you begin... even if only roughly. Then, with your new focus on collecting and analyzing data, you may find that some needles simply do not move, while others only a little. Be prepared to see very little ROI for an extended period of time.

The challenge for most lies in quantifying the economic benefits in terms of expected savings to plug into the ROI formula. Estimate benefits as the difference in expected costs of the status quo scenario vs. the predictive maintenance scenario. To calculate either, you need data covering: 1. probability of certain failures to occur within an interval 2. economic consequences of each failure (i.e. unplanned downtime, waste) 3. costs of the maintenance tasks involved (i.e. labor, data collection/analytics) It's unlikely you have all the data, so you'll have to introduce assumptions, which may appear shaky ("reduction of downtime by X% across the plant"), unless you become very specific about the assets involved and their maintenance.

1) Report and document the most significant cases of early detection of failures. 2) Calculate the overall cost made to get that detection (investment) and compare it to a "what if" in a late detection scenario (return). 3) A late detection scenario doesn't mean a "worst case" scenario... instead, I'd recommend the most probable output.

In over two decades in predictive maintenance across diverse sectors, I've seen its value often overlooked due to communication gaps. Effective PdM communication must be audience-specific: maintenance teams need actionable insights, operations seek reliability assurance, stakeholders look for financial returns, and the broader community benefits from shared knowledge. Success lies in translating technical achievements into compelling narratives for each group, backed by solid data and real-world examples. Bridging this communication gap not only showcases PdM's unseen value but also secures its place as a key strategic asset.

Leveraging basic Organizational Change Management (OCM) is important to the success of any program. Even stories in the monthly newsletter of things that did not go so well are really good to build credibility. People are much more willing to accept your stories of failure and recovery than your stories of how great you are.

- Create a dedicated area to share the most significant achievements. - Safety, cost savings, availabilities, and environmental care are the most frequent desirable outputs/objectives. Be clear about how the predictive maintenance strategy connects to them. - Celebrate.

A common oversight in predictive maintenance (PdM) is not treating it as a capital investment. Over 90% of PdM projects are funded from maintenance budgets, a practice I observed firsthand as a technology provider for vibration analysis and laser alignment systems. This end-of-year spending to avoid budget cuts leads to piecemeal equipment purchases without a strategic investment process, resulting in unused assets. My advice: even small PdM investments should undergo ROI analysis to ensure they meet strategic objectives and deliver desired outcomes. This approach mandates the establishment of clear procedures, goals, and metrics, transforming PdM from a cost to a value-adding investment.

- Keep learning from others. - Request and motivate revision by peers. - Don't be afraid to recognize mistakes. - Prepare yourself knowing how data management and AI can help in your predictive maintenance strategy.

Las empresas que buscan alta confiabilidad y disponibilidad de sus máquinas y procesos están convirtiendo sus programas basados en una estrategia de mantenimiento preventivo en programas con un enfoque proactivo buscando eliminar (no sobrellevar) los modos de falla de manera sistemática y monitorean las causas de falla y sus síntomas mediante el uso de tecnología avanzada. El mantenimiento basado en condición (MBC) incluye tanto el enfoque proactivo de identificación de causa de falla, como el mantenimiento predictivo de búsqueda de síntomas o efectos de falla.