Connected Service Parts Management is Not Predictive Maintenance

When explaining the concept of connected service parts management (CSPM), I always come across a misconception that it’s the same thing as predictive maintenance. Here’s why it’s not.

CSPM uses the same causal forecasting algorithms as traditional service parts management solutions. What distinguishes CSPM from SPM is that the former inputs data from connected assets into those causal forecasts. Those connected assets could be aircraft, MRI machines, servers, or whatever.

The difference between CSPM and predictive maintenance

As the name implies, predictive maintenance involves predicting which parts on a particular machine will break at any point in time. Once a service organization has this information, it can send a technician to fix that machine when it isn’t in use, before it breaks.

CSPM doesn’t predict which parts on which machines will break in a given time. Instead, it gathers four data points from smart, connected products:

  1. Where the assets are located.
  2. Each asset’s utilization rate (mileage, flight hours, scans per day, etc.).
  3. How many assets there are across the installed base.
  4. What the real failure rates are by model, age, application, environment etc.

It then takes all this data to model and forecast asset utilization over the next three months, six months, or however far into the future you want to go. Parts planners can then use those forecasts to generate parts demand forecasts.

Alright, well then why bother with CSPM at all? Because it eliminates the need to use proxies of how many assets are in your installed base, where they are, and how much use they’re experiencing.

To be fair, the Servigistics Business Unit currently is researching ways to institute a predictive maintenance-like capability. Until then, we’ll continue to provide parts planners with reliable, accurate data:

Watch IoT Parts Planning Webinar