One way to measure maintenance and evaluate productivity is to determine the amount of useful work produced per hour of input. Sounds straightforward, doesn’t it? In the field of maintenance, however, this is more complex than in most other fields.
If a technician is working efficiently, accurately and professionally to remove and replace a component on unit 34, we would say he/she is productive. The technician might have three hours of total input – including travel, parts pick-up and the work itself, and with that managers might feel satisfied.
But suppose the technician was given the wrong unit number: 43 instead or 34. How does that affect his/her productivity?
Another technician is working on the right unit with the right materials, but is doing a repair which will fail within two years rather than lasting 10 years. This technician planned better and works faster, doing the work in an hour instead of the allotted two. Is he/she productive? Suppose the repair lasted five years?
These are tough questions, and they have no easy answers.
Company Culture Counts
Our organization’s systems and benefits, along with the human resource and safety departments, all impose non-productive time on technicians, though some is necessary. Within the same organization, two maintenance departments across town from one another could each have very different initial productivity levels based on system-imposed activities – meetings, availability of tools, distance to the parts room, etc.
Does your organizational culture help or hinder the maintenance process? Every culture creates a pattern. The pattern either supports or defies the ability to preserve assets and provide capacity.
By way of example, in some governmental maintenance departments, the pattern created by the culture stands squarely in the way of an efficient process. A maintenance worker there has no authority to spend money to solve a problem, say to phone Grainger’s and go over and pick up a new hydraulic motor. They might have to endure a procurement process to get a more expensive, inferior part and wait weeks to boot.
In another case, a large field service company issued $50 bills to its technicians to help them solve any customer’s maintenance problem immediately. There was no call-in for authority, no second guessing. They just had to turn in the receipt and get reimbursed.
In both cases, the technician’s time is impacted by the systems, procedures and attitudes of their organization. This pattern can be seen by studying how technicians are forced to use their time.
Foundation Information
But first, maintenance leadership needs to answer some basic questions, including:
- Are there enough people for the work?
- Are they properly managed and supervised?
- Are they adequately supported?
- If the way they are supported were changed, could they do more work?
- What is the optimum capacity of the work group as now organized?
- Would extra training increase productivity?
Other basic information needed, which is usually available from the payroll department, includes:
- How many hours do you pay for daily?
- How much regular time?
- How much overtime time?
- On each shift?
Make a simple comparison between the payroll hours and the hours worked on maintenance jobs for a quick analysis of productivity.
Random Snapshots
Imagine taking random snapshots of your maintenance technicians. You would find that at the instant of the snapshot, a percentage of your crew is involved in marginally or non-productive activities. Work sampling is a formal technique for snapshotting to evaluate the activities of your maintenance workforce.
Studies show that 80 percent of productivity losses are directly attributable to management attitudes, systems and procedures. Only 20 percent can be traced to a worker’s motivation, attitude, energy or desire.
Most supervisors informally – or even unconsciously – do their own version of work sampling as they walk around observing things. It is far better is to have a formal methodology like work sampling. Basically, this is a technique for determining the proportion of time spent by workers in various defined categories of activity by making a number of observations in a random manner at random intervals of time over a particular time period.
Work sampling management can secure facts and uncover patterns about an operation without watching everyone all the time. It is systematized spot-checking, where different observers under the same conditions will get the same results.
Work sampling is important because:
1. It will help you see the real patterns present so you can attack the problems where they actually occur.
2. It lets you know how much lost time you have to work with.
3. You need a baseline of your shop’s productivity level before making changes.
4. After you change the culture or install some labor productivity improvement, you need to be able to measure the effect.
Problem Finder
Work sampling itself is not a problem solver. It is more of a problem finder. If properly planned, it will give very definite indications of what is going on and you can decide what should be done.
For example, a work sampling study may show that excessive time is spent waiting for materials in the morning. Judicious planning could allow parts room personnel to pull standard jobs the night before when the window is quiet. In the morning, more technicians can immediately get to work.
Once problems have been isolated, means can be selected to improve the situation. Work sampling can be used after the observations have started. Plus, you can keep reinterpreting the results for different uses.
Observations mount up rapidly. If you randomly observe a 20-person department six times a day for two weeks, you’ll have 1,200 observations.
Let’s suppose a 1,000-observation work study yields:
Activity Observed Number of Observations Occurrence Percent
All of the work categories 320 32
Talking 180 18
Travel 150 15
Unable to locate parts, tools 150 15
Waiting 100 10
Idle 100 10
Total 1,000 100
In this shop, what useful conclusions can maintenance managers come up with knowing that only 32 percent of the time, or 153 minutes, are spent working a day?
Initial Steps
Before conducting a work sampling study, there are some things that need to be done, including:
1. Define the scope of the study. It is advisable to begin with a general study to see how much time is spent in the general categories. At a later date, you might sharpen the focus and study particular lost time areas, such as parts waiting time.
2. Plan a study that will address the problem at hand and assign people to the study.
3. Review the study with the people in your crews. They may not like the idea, but they should be informed of their responsibility to contribute toward efficiency. Show them how high efficiency will improve their jobs and their quality of life.
After productivity improvements have been made, perform another work sampling study to determine if changes in procedures, systems, culture, tooling, parts room, supervision, etc., have actually increased the time spent working.
Keep In Mind
Once initiated, there are six things to keep in mind when conducting a work sampling study:
1. Randomness is the key to an insightful study.
2. Vary observation routes through the maintenance facility to increase randomness and help “surprise” your workers.
3. Train others to conduct the study and randomly vary the observer.
4. Select random times for observation. Plan four to six tours per day, allowing a reasonable time between tours. If the selected time runs into lunch or other established breaks, skip that tour.
5. A principle for success in work sampling is impartiality. Do not prejudge what you see or mix in outside factors – for example, if you think you know that someone is a “good” or “bad” worker.
You are trying to analyze the system, not the people. Do not focus on what the people were doing or what they were about to do. Concentrate only on what they are currently doing.
6. Record a person’s activity before you are seen by them.
Tally Sheet
When compiling your information, use one time tally sheet per shift (see the accompanying Sample Tally Sheet). Use a separate sheet for each observer.
Fill out the random times and random tour routes in advance. Enter them in the observation time and observation route rows on the sheet. Conduct the tours using the routes indicated at the times indicated.
Record the number of technicians at work the day of the observation. That is the number of observations you will make each tour. Enter that number in the available manpower row.
During each tour, record the observation elements – what is being done – in the column of the specific observation time. You can use tally marks as you walk through. Tally one element (check mark) per observation, per person, per tour.
If you observe crew members working in the wrong area or on the wrong job, make your observations where they are. Don’t judge where or what they are doing, just if they are working.
At the end of each shift, total the observations and transfer to this to a recap sheet.
An alternative to manual work sampling methods is the ISampler iPhone App, developed by Joel Levitt. It is a free, complete system with one-click reporting that enables quick and easy work sampling. To learn more, visit: http://www.work-sample.com.
Conclusion
In general, most maintenance managers have only a vague idea of how their technicians spend their time. By employing work sampling, the types of activities that technicians perform and the amount of time they spend on each activity can be identified. Armed with such information, managers can make decisions regarding allocation of technicians and activities to increase their efficiency and effectiveness.
With improved shop productivity, more work can be done in less time; that reduces vehicle downtime.
Realize, though, that productivity – which is also influenced by technician motivation and commitment – is not just a labor issue. Better tools and equipment, shop and work area layout and design, as well as improved technician education and training also can contribute to increased technician and shop productivity.
