[Cliff Williams] Wrench Time: What It Is, and How to Make It Measure Up

Many times I’ve been involved in strong discussions about the measure "wrench time" -- the measure of the time workers spend using tools on the job, performing maintenance. I’ve heard opinions that range from "totally useless" to "the best way to improve maintenance."

The questions you need to answer before you can decide which is closer to the truth, you need to understand the basics, which include:

• What is wrench time?

• How is wrench time measured?

Here's my explanation, with lessons learned from my own experience with measuring wrench time so that you too can measure it effectively in your organization

What Is Wrench Time?

Some people believe wrench time is the time shown on the work order. Others say it’s the time the trades have tools in their hand. However, both are focused only on tradespeople. The first example expects the tradespeople to account for their full week’s work, and the latter doesn't actually account for how to measure it.

When I suggest to the first that if they keep on measuring that way then they will quickly find that they are 100%, they seem pleased -- until I tell them that it won’t mean anything and have very little integrity. 

The Truth About Wrench Time

If you want your tradespeople to account for all of their hours and you consider more to be better, then it is fairly easy to get to 100% -- the time taken will expand to the time available. However, unless you are perfectly manned for the work available and have perfect planning and scheduling, there is no way that your tradespeople will be gainfully employed every hour of every day.

This form of measure will do what all measures do:  drive a behavior. The problem is that it’s the wrong behavior, because the tradespeople will stop entering the actual time taken for the work and just fill in their time, and they begin to think the whole system has no integrity, and that they are being recognized for doing the wrong thing.

The reality with that approach is that, even if the tradespeople were entering the actual total time they take to do the job, they are not measuring the traditional definition of wrench time. This sort of measurement won’t drive improved efficiencies -- it will only drive improved metrics, which is the last thing you want because those people will be recognized as "doing better."

The truth is that wrench time is not a measure of trades efficiency or effectiveness. In fact, if used properly, it’s not a measure of trades at all. Used in its widest sense, wrench time can be an powerful indicator of the organization’s culture.

How Wrench Time Should Be Measured

If you think about how wrench time is measured by those that actually have a good idea as to how it can be used, then what you will start to see are the reasons why tradespeople end up with their hands on the tools between 30% and 50% of the time (for the high performers.)

Wrench time should show things like the tradespeople waiting for equipment to be locked out or a permit issued, visiting stores a few times, and so on. But then planning and scheduling become front and center,  and you would consider it a measure of their effectiveness -- but is it?

My Own Experience With Wrench Time

Back in the early 1970’s -- before many of you were born -- I had just finished my apprenticeship and taken a few courses at the local college, including the fundamentals of Industrial Engineering. I was working at a plant belonging to one of the nationalized large organizations in the U.K. At the time there were more than 150,000 people working in the national steel industry -- about 3,000 of them at our plant. Out of that 3,000 there were about 280 tradespeople -- mechanical, electrical, instrumentation, electronics, and a smattering of the civil trades -- painters and carpenters.

It was well understood back then that the efficiency of the trades workforce was absolutely terrible. Most of them were simply working in a reactive mode. When they got a call, they fixed it -- seldom with any record of what was done when, or on which piece of equipment. Given the size of the plant, there was invariably one production line down every day for bigger repairs, time-based change outs, and the odd inspection. Overtime was rampant. We would even bring in an extra supervisor to help, and yet the work was simply a hand-written list with work orders being handwritten also – and issued but never returned.

Then Along Came Big Change: Wrench Time

Someone, somewhere decided that this couldn’t continue or, most likely, a consulting company based in London got the ears of those in authority and told them that it was terrible but they could fix the problem. These consultants showed up at our plant wearing suits and talking with "proper" accents. They told us about what they intended to do and what was required of us -- they were going to roll out a Computerized Maintenance Management System (CMMS) and we were going to use it to give feedback. 

Now, as you can imagine, the only thing that most of us knew about a computer in the early 1970's was that it lived in the "computer block" and it took an army of people to feed it! These consultants were smarter than we expected. When they said that the feedback piece would include the time taken to do the job, they were greeted with jeers about "big brother" and the older trades guys simply laughed at them. But they got everyone’s attention when they said there would be a bonus paid for it.

There were numerous meetings after that, and it was explained that they were going to set up a system in the big mainframe computer that would print out work orders -- some on a regular basis, and some as the work cropped up. They were going to set times for the jobs and we would get a bonus based on how close we came to completing the work in the time allowed. The obvious question from the old-timers: "And where are you going to get these times from?"  To which the consultants replied, "We are going to do time studies."

Time Studies in Action

Time studies set the place alight, with talk about time and motion and stopwatches, and where the consultants could store these stopwatches started -- especially with the trade union officials. Local management stepped in and told everyone that this was going to happen, not just at our plant but across the whole organization. We were the lucky ones, as we were one of the first, so we’d get the chance to challenge what was happening and fine-tune it, so it could be rolled out. It seemed we had no choice.

The next phase was when the consultants asked for a number of tradespeople to be seconded from the shop floor to help them out -- but the way they put it was that they wanted union representation to ensure the integrity of what was being done. A few of us younger tradespeople looked at the opportunity closely. It meant steady day shift, no grease in your hair, and what seemed to be a fun time.

Even though the union wanted representation from some others more militant than the young tradespeople, nobody was prepared to give up their security blanket of doing what they had been for many years. This meant that they had to reluctantly accept that the younger guys would be the ones who worked with the consultants. This was at a time in the UK when trade unions had immense power and a situation like this was a politically-charged time bomb. 

Anyway, I’d enjoyed the topic of work-study when I had taken the Industrial Engineering course and so I was pretty happy when I was chosen to become a "Planner Categorizer." The consultants explained that there would never be enough time to study every job at the plant, and so we were going to study representative jobs. When we had enough of them, we would then extrapolate and set the times for similar jobs, categorize them, taking into consideration complexity and environment. This led to us being trained in time study, where we would watch a task being done, note what was happening, change the description at every task that was less than 30 seconds, and just reset every 30 seconds if the task was longer.

Wrench Time as a Rate

We were introduced to a concept called "rating." The way they explained it was that a sprinter could run as fast as he could for a 100 yards or so, but never for a mile or a marathon. What they were looking for was a rate, or speed that a person could work at for 8 hours -- taking the appropriate breaks, etc. -- not rushing and not deliberately working slowly.

Rating obviously didn’t get to the shop floor, because when we did eventually start to study the jobs, the tradespeople would run from place to place or deliberately work in slow-motion just to try and confuse us. The consultants described this as "standard time," and that’s what the duration of the job would be -- not the actual hours and minutes that the job took. The second and more time-consuming part of the exercise was the analysis of the study, and this usually took about 4 hours for every hour studied.

During the analysis, we would multiply the time taken by the rate assigned. If someone was rated at a 90% and they took 30 seconds to do the task, the time was calculated as 27 seconds. The more important part (though we didn’t realize it at the time) was the task of identifying whether or not the task was valid work time. By that, the consultants meant the time the tradespeople were using tools to do the job.

Any time that didn’t fall under the "effective work" code was coded as "non-productive" time and split out into various categories -- going to stores, waiting for lockout, having to fit components such as clevis pins etc, standing around talking, seeking information and so on. These times were removed from the "actual hours" of the study leaving what was then considered the "standard time" for the job, although there was an allowance applied as it was recognized there would need to be some recovery time if a person was working at this rate for 8 hours. 

The fact that the tradespeople’s bonus depended on them returning the work orders with information, resulted in us getting some excellent data. When they found that the time assigned was only for the "effective work" portion, they very quickly told us when they had to wait for parts, lockouts etc. so that they could get credit for that time.

The Key Lessons Learned

Here are some the key lessons this taught us.

1. Get Visibility Into Your Organization

With a couple of years worth of data, we could see that the tradespeople were actually spending about 30% of their time doing "effective work," and we had the break out of the remaining time in the various categories of "non-productive" work. By this time, the consultants had stopped visiting as they really were only contracted to set up the system, so using the information was now down to us. At this time, "secondment" had ended and the few of us who had remained were now the planning department.

Armed with the data that showed the team was wasting time waiting for a variety of things, we set up meetings with the plant management team, who were unsurprisingly quite upset with what we showed them. They decided that we should cascade the information down through the department heads through to the front-line leaders, and even include the trade unions. They said that they would be there support us at each step.

When we made the presentations, every department had excuses for why the results were the way they were -- we didn’t have enough people to do the lockouts, there was always a line up at the stores so people had to wait, even to the point where some of the trades supervisors told us that they instructed the tradespeople to wait for them before making decisions.

2. Take Accountability

It took a bit of courage for us in the planning department to admit that we played a major part in the problem, and when we said that we didn’t really communicate clearly and early enough, everyone else seemed happy to let us be the fall guys. They weren’t so enthusiastic when we said that we could easily remedy that with a little input from them, but after that it would be their responsibility to communicate and translate it into lower, or even zero, waiting time. 

We set up planning meetings for every afternoon. At these meetings, we would review the planned work for the following week’s shutdown. Operations and front-line supervision would attend to raise any concerns. We promised we’d get the finalized list to the department heads and supervisors by the following day, so that they could plan the labor requirements. We also promised to send a list of work orders with the replacement units identified to the stores group, so that they could check that we had stock and maybe pull them off the shelf and put them somewhere safe. 

3. Improve Wrench Time Through Trial and Error

We did this for a little while but didn’t see the results that we thought we would, so we used another tool that I had learned at college and that was "activity sampling." We reviewed a group of people and noted what they were doing in general terms every few minutes. We saw that even though lockouts were known and taking place in a more logical fashion, the majority of people were standing around and there still was a line up at the stores taking time.

When the days progressed we also saw that parts still needed to be fitted and people were searching for supervisors for decisions. This led to the second round of meetings where, faced with the facts, everyone agreed that we needed to improve in other ways. What came from these meetings was that electricians and operators started the lockouts prior to the shutdown shift, parts were delivered to the job site the day before -- after making sure that the clevis pins would fit in the clevis. Instructions were rewritten so that tradespeople didn’t need to consult the supervisors, and the supervisors took it one step further by meeting with as many of the tradespeople as they could, the day before shutdown.

4. How Wrench Time Is an Indicator of an Organization's Culture

The result of all of this work was that for the process line on shutdown the "effective work" reached the high-70s%, the tradespeople were a lot less frustrated, and the supervisors, who at first thought we were trying to usurp their authority, asked why we hadn’t done this sooner. That was all in place in 1976, and by the time I left for Canada in 1981, we were scheduling work for the "run" shifts at about a 40% level. Tradespeople were creating written work orders and then we would load them into the computer, we were now running a backlog.

Overtime had been reduced, but people had become used to it and actually appreciated the efficiencies in its place.

However, the biggest change was in people’s attitudes. Everyone had become more collaborative, everyone felt part of the process, everyone gave their opinions and everyone listened to those opinions.

Now it isn’t to say that things were perfect -- there were still one or two holdouts but more and more their peers were holding them accountable -- but we had been successful. We saw success in improving the true meaning of wrench time, success in changing behaviors and so changing attitudes and so, without realizing it, and success in changing the culture of the plant.

Wrench Time Is Not Just For the Maintenance Department

Obviously the first thing is to recognize is that if management is using wrench time as a measure of only the maintenance department, or of maintenance worker performance, they are making a big mistake.

Wrench time is quite often the most divisive measure used at a plant. Without clear communication, it can very easily feel like people judging the efforts of the maintenance team, when the team has little control over the results. If the organization wants to use wrench time to its most effective end, then they need to make it understood that even though they will initially be measuring what the result is, the benefit will come when they understand why the result is what it is, and then actually doing something to improve it, making it clear that all departments could be involved

How to Measure Wrench Time Effectively

Once everyone is clear that wrench time is a multi-functional measure and that it is designed to drive improvements and not to assign blame, there are some simple steps to take.

1. Use a CMMS to Identify Both Wrench Time and Wait Time

Whereas all of those years ago we had a very basic CMMS and the only way available to us to determine ‘non-productive time’ was by carrying out time studies – that is not the case today.

Most, if not all, CMMS have fields available to enter time spent waiting or delayed that can be used to identify this ‘non-productive time’. This field needs to be populated with the types of non-productive time that you experience at YOUR organization. For example, if a permit is required before starting work, then waiting for it should be listed. Additionally, you should list if lock-out-tag-outs are carried out for the task. To make the best use of this field, you should include the less typical issues, such as waiting for information, waiting for permission, waiting for help, as you will find that these problems are more indicative of the cultural challenges you face.

2. Explain Why You Are Making an Organizational Change

You may find that getting people to take the time to provide these details is not quite as easy as it sounds and that is usually because the change isn’t being managed correctly.

People have to clearly understand why the change is necessary, and great care must be taken to ensure that no findings are portrayed in a negative light, as one negative can derail the whole initiative. Once you have everyone (or the majority) on board, then you need to tabulate the findings and present them to the whole group.

The charts or data can be used to provide different perspectives. One chart may list the number of times a particular "wait" occurred, one may show the number of hours lost to that "wait," and the one that will get the most attention would be the amount of money lost. This could include value of lost production, as well as the wages of those standing around waiting

3. Analyze the Data

It’s at this point that the gains can be made. While analyzing the data, you should be looking at who actually controls the delay, who influences it, and how they can reduce or remove the wait. What should happen at this stage is that there should be small working teams breaking off in an effort to develop the solutions that may involve breaking down some widely held paradigms.

It’s always a good practice to bring the solutions back to the bigger group before implementing, so that you can confirm that one solution doesn’t conflict with another solution. Periodic review or trending will help you ensure that things don’t start to drift.

Is Measuring Wrench Time Worthwhile For Your Organization?

If done properly, measuring wrench time will provide some significant rewards. If done incorrectly, it can often cause further problems and create more divisions than already exist.

The key to making it successful is remembering what is the beginning, and why will point you at the opportunities, and identifying those with the authority and tools to realize those opportunities will give you the results you desire.

Wrench time has long been thought of as a good way of identifying the "invisible workforce," and allow you to be more effective and efficient. But it’s even more than that: It is a great tool to change your culture, and allow your organization to be the best that it can be.