How to collect the failure data needed to drive improved plant performance.

The first step to any reliability improvement program is to define what data or information will be required to drive defect elimination.  In the previous posts, the following topics were discussed;

All of the topics discussed what data is required and how to structure the data to implement a reliability engineering program or defect elimination process.   Understanding the what, when and why set up and is required to setup a reliability engineering program.  However, where many organization fail in this process is getting the data collected in a consistent and timely manner.

One of the hardest parts of reliability engineering program is the collection of the data by the technicians in the field.   So how can an organization ensure the data is collected consistently and promptly?

Prior Preparation Prevents Poor Performance

In order to make the collection of failure data easier, the method of collection must be well thought out in advance.  In order to simplify the collection of failure data, a Master Library System should be utilized.  A Master Library system is a system that ensures that the object codes are specific to the asset, the damage code is specific to the object and the cause codes are specific to the object and damage codes.   This approach ensures only relevant options are displayed in the CMMS (or whatever type of system is being used.)

The Master Library has a list of objects, damage, cause codes and other codes that are identified with a unique identifier (e.g. OB027=Bearing).  These codes are then mapped to specific asset classes & types.   This is why building a proper asset hierarchy is so important.   By utilizing this approach, the number of unique codes can be kept to a minimal while ensuring only relevant codes are available to the end user.

Obviously, to develop this type of library and connections is not a simple task.  But by taking the time upfront to develop these relationships, the end user will have a much simpler and easier experience to capture the data.

After all, the goal of any data collection system is to be simple to use and understood by the front line maintenance and operations staff.  The system so be simple enough to use, but detailed enough to provide the right level of granularity.  Also, the failure codes should be mutually exclusive, which means that a bearing will only ever have one code, but could be used in multiple lists.

An example of a Master Library system in action with a complete code structure would be Asset-Part-Damage-Cause-Trade-Work Type-Consequence.  When populated the failure code would be Wheel-Nut-Looseness-Incorrect Installation-Mechanical-Unplanned-Safety.  Here you can see exactly what the issue is, who addressed it and the consequences of the failure.

Make It Easy

With the Master Library of failure codes established, the data collection process will be a bit more revelant and simple to use for the end user.  However, there are a few additional things that could be implemented to ensure the data collection goes smoothly;

  • Ensure the failure data can be captured close to the incident.  Whether this is through the use of a mobile solution or by have terminals strategically located throughout the facility.
  • Keep drop down lists as small as possible.  The lists should not require any scrolling and ideally kept to less than 10 codes.
  • Involve the end users in the development of the structure.   Since they will be the ones entering the data, allow them to try it and provide recommendations to improve the collection process.

But making the system simple and easy to use is not enough.  You must gain the buy-in and support from the end users.

Create Buy-In, Train, and Coach

Any new activity or change being implemented will not be as simple as handing over a new process.  There are certain steps that will be required to ensure the frontline staff not only perform the data collection but understand the importance of having accurate, timely and consistent data.    So what are the steps required to engage the end users?  Let’s look at applying the ADKAR model to this situation;

  • Awareness – Create awareness with the end users that either the organization will begin to collect failure data or will be revising the method.
  • Desire – Communicate the reasoning for the change and focus on the What’s In It For Me (WIIFM).  The end users need to understand the benefits to them for this new process.
  • Knowledge – Provide training on the new process.  Do not just focus on the how, but also be sure to address the what and why.
  • Ability – Have the end users practice the new knowledge through simulations or other training activities.
  • Reinforcement – Reinforce the good behavior observed.  When someone takes the time to use the new process, or ensure that accurate data is used, be sure to thank them.

With all of this in place, your organization should be in a good place to begin collecting and using the failure data to improve plant performance.  Does your organization utilize a Master Library approach to failure coding?  If so, how successful has it been?  What are the benefits that you have seen as a result?  If you don’t have a Master Library approach, how successful or simple is your failure coding?

Remember, to find success, you must first solve the problem, then achieve the implementation of the solution, and finally sustain winning results.

I’m James Kovacevic
Eruditio, LLC
Where Education Meets Application
Follow @EruditioLLC