It is said that if you cannot measure something, then you cannot manage it (or improve it). If that is the case, then how would we measure the productivity of our Maintenance Departments (or our maintainers)?
For organisations in many asset-intensive industries, maintenance costs can be a significant proportion of total operating costs (perhaps as high as 50% of total costs in some parts of the mining industry). In addition, maintenance can have a significant impact on other aspects of organisational performance, such as safety, environmental and production performance. It is, therefore, highly important that maintenance operates in as productive a manner as possible.
In this article, we will reveal that measuring Maintenance Productivity is not a simple matter, and there is no single measure that can easily be applied. Nevertheless, we will give you some guidelines and tips which will help you to assess, and improve, the productivity of your maintenance function.
What is Productivity?
Before considering maintenance productivity, let’s first define what productivity is.
In general, productivity is considered to be the ratio of outputs to inputs of a system. Outputs typically consist of products or services (or the value of these products or services), and inputs are the resources consumed (or the cost of those resources consumed) to produce those products or services. Because productivity is a ratio, it is important that both the numerator and denominator of the ratio have the same units of measure, and typically this unit of measure is financial (Dollars, Euros, Ringgit or equivalent).
The other key concept to be aware of is that, because productivity is a ratio, being productive requires a focus on both effectiveness (doing the right things) and efficiency (doing things right). Being effective typically will increase the value of the numerator (outputs) in the productivity equation, while being efficient will typically reduce the value of the denominator (inputs).
What is maintenance productivity?
Applying this general definition and concept to maintenance, it becomes clear that we need to be able to define (and measure) both the outputs and inputs associated with the maintenance system.
Inputs to the maintenance system are comparatively easy to define and measure. The resources consumed in performing maintenance typically will consist of labour, materials, tools or equipment used to maintain the plant. Financial values can be determined for each of these things, and typically would be measured in most organisations (although some accounting jiggery-pokery may be required in order to ensure that an appropriate level of corporate overheads is applied to the maintenance function – for example for the cost of owning and operating workshop buildings, or for providing corporate IT support to the maintenance function).
What is the value of maintenance?
It is the numerator in the productivity ratio that is more problematic. How do we value maintenance outputs?
The outputs (or benefits) of a maintenance function would normally consist of one or more of the following:
- Plant or equipment uptime
- Maintenance cost avoidance
- Operating cost efficiency
- Product Quality
- Environmental and/or Safety risk reduction
Let’s examine each of these in a little more detail, and discuss how we may value each of these benefits.
Plant or equipment uptime
Probably the most important function of maintenance is to deliver equipment capability – the capability for plant and equipment to fulfil its intended functions; those functions that are required in order to meet organisational output requirements. The real question is how do we value this, in financial terms? What would equipment capability be, if maintenance was not performed? Over what time period do we assess this? And what is the incremental cost to the organisation of any shortfall in output?
One of the challenges in assessing the value of maintenance is that there is very often a time lag between the absence of maintenance and any noticeable drop in equipment performance, maintenance costs or uptime. And this time lag can often be quite substantial. So it may appear to the uninformed that cutting maintenance costs has improved maintenance productivity, but they may well find that they pay the penalty for the absence of maintenance (in terms of equipment performance or costs) in future time periods.
Further, how do we value the cost of “lost” production? Is it the additional marginal profit that would have been achieved if the product had been produced? Do we include allowances for overheads and other fixed costs? What about non-cash costs, such as depreciation? What if the revenue from this “lost” production has not actually been lost forever, but simply deferred to a future time period?
The answers to these question are not simple, and there is no standard way of answering them. This makes it difficult to truly estimate the value of maintenance, and therefore maintenance productivity.
Maintenance cost avoidance
Spending a small amount of money on (the right) maintenance at the right time often avoids the need for much greater expenditure later. For example, spending a small amount of money on repainting a building or structure avoids the need for much more extensive and expensive repairs in future. This clearly is a valuable activity – but how do we quantify its value? Again, not an easy answer.
Operating efficiency
Maintenance is often directed at ensuring that equipment operates efficiently. For example, turbine blades are periodically cleaned to maintain turbine efficiency. Soot blowers are maintained in coal-fired power stations to ensure boilers operate efficiently. The value of these activities needs to be estimated as well, if maintenance productivity is to be measured.
Product quality
Well maintained equipment allows organisations to ensure that its products (or services) meet quality specifications. Local councils mow the grass on parks and playing fields in order to ensure that they meet the expectations of their local communities. Lathes are maintained in order to ensure that the items that they produce meet sizing specifications. If this maintenance was not performed, what would be the financial impact on the business? This is particularly hard to measure where the asset being maintained is providing a service (such as, for example, a local park) rather than producing a product.
Environmental and / or safety risk reduction
And finally, maintenance is often performed in order to reduce or mitigate risks associated with equipment failure. In some cases, failure to perform this maintenance could result in a failure to meet the conditions of operating licenses, and could, therefore, result in the entire operation being forced to close. In other cases, it simply reduces the overall risks to employees and the public. In order to be able to calculate productivity, we need to be able to express this reduction in risk in financial terms. Is this possible? Certainly, however it is not easy. To do this, we need to estimate the likely impact of the event in financial terms, and we also need to express the likelihood of the event reasonably accurately in quantitative terms. It is this second element that is the most difficult. This is particularly the case when the event has an extremely low probability of occurrence, as there will be little statistical data available from past events to use.
So where does that leave us?
While estimating the value of Maintenance is certainly possible, it is not easy. And therefore, not really practical – at least not on an ongoing basis. So instead, we need to rely on other ways of estimating maintenance productivity.
Tool time / wrench time as a measure of maintenance productivity
Tool time (or wrench time, as it is better known in the US) is often used as a proxy to assess maintenance productivity. This involves performing time studies on maintenance tradesmen and technicians to determine what proportion of their time is spent with “hands on tools” and what proportion of their time is spent on other activities. The most common ways of performing this measurement are either to:
- Have an observer follow a maintenance technician around for a period of time (usually several days or weeks) and use a stopwatch and clipboard to note the activities performed by the maintenance technician and the time spent on each activity, or
- Use a technique called Activity Sampling or Work Sampling (see https://en.wikipedia.org/wiki/Work_sampling) to observe maintenance technicians at statistically generated random intervals and note the activity being performed at that specific point in time. The number of observations to be taken in order to provide a given level of confidence in the results can be calculated.
There are several issues associated with using these techniques.
- First, if people know that they are being observed, then they can (and probably will) modify their behaviours – at least for a short period of time – and so the results may not be representative of longer term behaviours.
- Second, there can be significant resistance to applying these techniques (it reeks of “Big Brother”) and in unionised or militant workforces, you may not be permitted to use them.
- Third, if applying Activity Sampling, the first thing that you must do in order to observe what a Maintenance Technician is doing is find them – and in larger workplaces this may be difficult. Especially if they do not want to be found!
But the biggest issue with using tool time as a proxy for Maintenance Productivity is that it only measures one side of the productivity equation – the inputs. It assumes that tool time, and only tool time, is valuable, and that the more time technicians spend working on the tools then the more productive the maintenance function will be. Time spent analysing and solving recurring failures so that they never recur would be considered (incorrectly) as a “waste”. Furthermore, time spent on the tools doing the “wrong work” (such as, for example, performing an unnecessary and intrusive preventive maintenance inspection that actually increases the risk of premature equipment failure) would be considered to be “productive” when it clearly is not.
Nevertheless, tool time studies can be valuable in providing some insights into the things that cause delays and prevent maintenance technicians from performing work efficiently. Time spent waiting for permits, waiting for instructions, waiting for parts, travelling unnecessarily between the workshop, warehouse and job site can all be measured. This can then be used to identify opportunities to reduce travel time and delays (for example, through better maintenance planning, scheduling and preparation) and prioritise these for action. However, you may be able to achieve this without the need for stopwatches and clipboards. How about asking the technicians themselves to keep notes on the delays that they incur? And involve them in workshops to identify and implement improvement opportunities?
Using standard times to assess maintenance productivity
Another possible approach that could be used to measure maintenance technician productivity is to compare the actual time taken to perform planned maintenance tasks with standard estimates for performing these tasks. If the actual time taken to perform these tasks is getting lower (in comparison with the standard estimates) then you know that maintenance productivity is improving.
The challenge with this approach is in ensuring that the standard estimates of task duration are realistic and accurate. In practice, in my experience, in most industries, there is little science applied to estimating the duration of maintenance tasks. Planners usually base their estimates on “gut feel”, and in many cases, often over-estimate task duration so that key performance indicator targets (such as Schedule Compliance, % Planned Work Completed etc) can be more easily met. Further, there is often pressure on planners to increase task duration estimates when technicians consistently fail to achieve the estimates (rather than the pressure being on the technicians to explain why they cannot achieve these estimates).
There is a range of alternative techniques that are possible for estimating maintenance task duration. Chapter 10 of Don Nyman and Joel Levitt’s excellent book “Maintenance Planning, Scheduling and Coordination” outlines some of these, including:
- Using Published Standards
- Gross Estimates
- Historical Averages
- Adjusted Averages
- Analytical Estimates
- Job Slotting and Labour Libraries, and
- Universal Maintenance Standards
Some of these are more time-consuming (but more accurate), but whatever technique you use, it should be consistently applied if you want to use standard job times to assess maintenance productivity.
Once again, however, this technique does not consider whether the work is actually “the right work”, and may lead you to simply do “the wrong work” more efficiently.
A balanced view of maintenance productivity
We can see, therefore, that there is no single, easy way of measuring Maintenance Productivity. What is required is a balanced set of performance measures that assess:
- Maintenance Effectiveness, and
- Maintenance Efficiency
In terms of Maintenance Effectiveness, we need to measure whether Maintenance is delivering the outcomes that the organisation needs in terms of:
- Plant or equipment uptime – the level of uptime should be sufficient to meet the organisation’s needs and no higher
- Maintenance cost avoidance – this may be difficult to measure, for the reasons mentioned earlier, but a possible leading measure could be the % of the Preventive Maintenance program that is delivered on time
- Operating cost efficiency – probably best limited to specific areas of operating cost where maintenance has the greatest impact (e.g. fuel efficiency or reagent usage)
- Product Quality – this could possibly be measured in terms of levels of production rework, scrap or reprocessing, or the proportion of product that is produced within quality specifications first time (first pass yield)
- Environmental and/or Safety risk reduction – this is most easily measured in terms of regulatory compliance, but this, on its own, is not a comprehensive measure of risk reduction
In terms of Maintenance Efficiency, the most suitable output measure is performance against a zero-based budget (either in total costs or cost per unit of output), but there are other leading measures of maintenance efficiency that may be useful, such as:
- Percent of maintenance work that is planned
- Levels of Maintenance rework
A comprehensive review of Maintenance Performance measures is outside the scope of this article, but you may find the eight guiding principles listed here to be useful.
Conclusion
As we have seen in this article, measuring Maintenance Productivity is not a simple matter, and there is no single measure that can easily be applied. Nevertheless, we have attempted to give you some guidelines and tips which will help you to assess, and improve, the productivity of your maintenance function. We would love to help you improve Maintenance Productivity in your organisation. Please contact us if you would like to give us that opportunity.