My Maintenance Program Cost Me!!

Facilities Management Portfolio Marketing Leader, IBM

If you’ve ever spent more than 15 minutes with me you know I’m a huge recreational fisherman and compete in many tournaments each year.  I routinely perform maintenance on my equipment: boats, rods, reels, knots, lures, hooks, electronics, etc.  But despite my efforts, poor maintenance cost me $5,000 last weekend.  My safety harnesses on my high speed trolling rods and reels failed and both of them jumped out of the rod holders in big seas and are now resting comfortably in 180’ of water.  THAT HURT!

If I had noticed the frayed safety lines I would have replaced them…an example of Reactive Maintenance.  A better method of maintenance would have been to replace the safety lines periodically; like every year…an example of Preventative Maintenance (PM).  Installation of sensors to detect and predict conditions (sunlight, saltwater, pressure) that cause failure would take it to the next level….an example of Predictive Maintenance (PdM).

Manufacturers have embraced reactive and preventative maintenance processes in their operations since the beginning of manufacturing.  PdM is a hot topic in manufacturing and is gaining traction as new products to enable these processes are released.  IBM’s Predictive Maintenance and Quality (PMQ) product takes PdM to the next level.  IBM PMQ delivers 5 major enhancements to traditional PM procedures:

1.  Capturing Data from Sensors on the plant floor - As more equipment is instrumented and interconnected, large quantities of data flows can be used to monitor, in real time, conditions that may indicate a potential failure.  Connecting to multiple sources and types of data are crucial to PdM.

2.  Predicting Failures – data from sensors is continuously scored using predictive analytics software.  Predictive analytic models mine the data and correlate past failures using multivariate analysis.  The models can mine all the variables and conditions that contributed to past failures in order to predict future failures.  Incoming data from sensors are then run through the model and asset health scores are generated on a real time basis.

3.  Dashboards and Reporting – monitoring and alerts to out-of-tolerance conditions or predicted failures can be delivered to operations personnel for proactive management.  Dashboards can be delivered at a company, regional, plant or asset level to provide insight into asset health or potential downtime.  Drill down in to historical and statistical reporting provides root cause analysis to drive better understanding of failures, productivity and costs.

4.  Utilization of Decision Management systems – decision management software is utilized to automate the decision making process.  Is the asset health score so severe that senior level technicians must be deployed in order to prevent a failure or downtime in the plant?  Or is the health score in a range that a simple inspection and a routine maintenance can be performed by an apprentice?  With the continuing loss of senior technicians due to baby boomers retiring from the workforce, decision management systems can capture the wealth of knowledge these leaders possess before they walk out the door.  And with the massive amount of assets being monitored, automating the decision making process just makes good sense.

5.  Automation of Work Orders – So we’ve monitored the asset, captured the data, used predictive analytics to predict failure, automated the decision making process…so how do we now automate the execution of the task of fixing or preventing the problem in the plant.  Most manufacturers have implemented Enterprise Asset Management (EAM) systems, such as IBM Maximo, in their plants.  These are the systems of record for plant equipment, spare parts inventory, plant work orders and asset life cycle management.  For instance, automation of the decision management process tells us that we need to deploy technicians to maintain the asset in question.  Providing an automated link to EAM and the creation of a “hot” work order minimizes the lag time from the prediction of the failure to deployment of human resources to correct the situation.

So what are the benefits of implementing PdM?

  • Lower equipment downtime – results in more throughput, better customer service levels and increased revenue.
  • Better utilization of scarce maintenance resources – deploy maintenance resources to critical and predicted equipment failures rather than performing preventative maintenance that may be premature and unnecessary.
  • Reduction of spare parts inventory – by better predictions of machine failure, maintaining a large safety stock of spare parts can be optimized.
  • Well maintained production equipment typically produces better quality products resulting in lower warranty claims.  This could potentially result in lower warranty reserves required.

I challenge you to respond to this post with other ideas on the benefits of PdM.

Ok, enough for now...I need to go to the tackle shop to replace my equipment.