I

n one recent reliability trade magazine, there were 34 advertisements for new and improved tools or training for improving your reliability program. The tools included software, new condition monitoring gadgets, and training and certifications on best practices. Further, the magazine included eight articles on such topics. However, there were zero articles and zero advertisements on the biggest lever by far for rapidly and sustainably improving equipment reliability, cost, quality, safety and throughput – shop floor observation.

With shop floor observation, there is nothing to buy. But by leaving this tool out, vendors are misleading countless practitioners on what really gets a reliability program up and going. Longtime engineers, managers and plant managers will tell you that nothing comes even close to the impact firsthand observation has on operations, yet no one talks about it. Instead, vendors invent new tools to keep you in the office tied to your computer or smartphone looking at biased data with intoxicatingly beautiful graphics. This, coupled with the fact that you are already overwhelmed with e-mails, conference calls and daily management meetings, has resulted in nearly zero firsthand observations of reality.

Data and metrics are biased. All data is subject to filters, manipulation and interpretation, yet most people swallow it hook, line and sinker as the truth. Data granularity is also lost when you lump all numbers together and take averages and totals. Let’s look at three examples.

Preventive Maintenance (Pm) Percent Compliance: Did you know that you get the same “credit” for the monthly inspection of fire extinguishers, ladders, extension cords, ground-fault circuit interrupters, and drinking water fountain filters as you do for lubricating the bearing on a critical production center? It’s true. A plant can be at 99 percent Pm compliance while not lubricating a single bearing in the plant, giving the leadership a false sense of reliability

Predictive Maintenance (PdM) Percent of Total Work Hours: Full-time PdM technicians are routinely pulled off their planned jobs to support emergency work and outages. As high as 80 percent of the time, these technicians are not doing PdM work. Supervisors who are strapped for resources to support the urgent emergency work quickly pull these strategic work resources. Hard to blame them though, they are often the highest trained, hardest working and passionate workers. Nevertheless, the metrics nearly always credit these technicians for 40 hours of PdM work. This practice also greatly understates the amount of unplanned work.

Schedule Compliance: It is common to audit an outage and find some discovery work has changed the critical path. Resources are moved off of less critical jobs to complete the work. Consequently, these less critical jobs are not fully completed, and since they will be on the schedule again for the next outage, they are closed out as complete. The raw data and key performance indicators (KPIs) reflect 100 percent completion.

These are not hypothetical examples nor extreme cases, but rather common ones. No one is trying to be malicious; most often, they believe they are doing, in the moment, what is best for the company. Regardless, the metrics are not accurate. What decisions are the leadership team making with this bad data? Are they working on root causes with faulty PM data? Are they thinking they are better than they are with PdM, yet failures continue to happen? Bad data leads to bad decisions, plain and simple.

But, not all is lost. While KPIs are biased, they are not worthless. They actually provide good insight into what to go and see via firsthand observation. Observation, however, does not mean the five-minute snapshot. It refers to the chalk circle observation exercise pioneered by Taiichi Ohno, considered the father of the Toyota Production System that inspired lean manufacturing. Strictly speaking, chalk circle observation is one in which the observer stands in a small imaginary circle on the shop floor and observes the seven forms of waste and standard work. These observations are at least four hours in duration and are best when they occur over multiple days. By doing such a sustained observation, the leader gains insight into simple changes that can prevent waste and lead to new action plans. If done correctly, most actions can be completed in 30 days. Imagine the impact on both the organization and sponsors if for zero dollars you produce significant results in just days.

Here are three examples of chalk circle observations that lead to actions that deliver impressive results.

Full Job Kitting: If you are tracking kitted jobs, you will most likely see that only the major parts are actually kitted. For example, the job is to replace a pump. Ten times out of ten, only the pump will comprise the kit. The planner assumes it is insignificant for the craftsperson to find a coupling, shim stock, gasket material, or any other shop part. Observed in practice, a fully kitted job often increases wrench time by as much as 50 percent.

Staging of Parts and Equipment: At the beginning of a shift, there is a mad dash to find a fork truck to move parts or to try and locate a 10-ton mobile crane and a welder. Often, one fork truck is shared among five teams of craftspeople. The crane was not left where it was supposed to be (it’s in the shop getting its monthly PM) and the welder is not working due to a damaged cable. This scenario is more common than you may think. You act and create a kitter/stager position to stage all the parts and equipment at the job site during the previous shift. By doing so, 100 percent of the morning chaos can be prevented, impacting wrench time as much as 50 percent.

Standard Work for Taking Crafts Off Planned Work: Your PdM technicians begin work at 7 a.m. on Monday and are called to support an emergency job until 2 p.m. With chalk circle observation, you see that the work they were called off to support was critical, but those crafts assigned to do emergency work were actually working on a job that could wait at least 24 hours. The supervisor, in an effort to please, pulled the team of people working on preventing next month’s emergency work. A simple fix is to create an approval process to pull these resources into the urgent work. By seeking second level approval, you can realize at least a 90 percent reduction in these events.

Note how simple, low or no cost, and impactful these changes will have on results. Also, reflect on the ability of KPIs to highlight these wastes.

Without a doubt, tools, training and certifications are critical parts of a reliability plan best serving those organizations well into their reliability journey. However, they are not a substitute for your most impactful action – observation. Further, highlighting only the cutting-edge tools and training is misleading those in the early phases of their reliability journey. Tools are not a substitute for firsthand observation.

Is this disease in your plant? In your next meeting, keep track of talking points that are based on biased metrics/KPIs, biased opinion and firsthand chalk circle observation. Expect to be amazed.