As I’ve discussed reliability with many folks around North America, I continue to hear about the difficulty in getting buy in for a healthy, aggressive reliability program to be put in place. So if it’s so important, why is it so difficult?

The trick continues to be getting senior management to understand what reliability is and what it can do for bottom line corporate performance. Since it’s not easy to get the ear of senior folks much of the time, there is another way to consider going about this…

If you can identlfy an asset or process in your area of responsibility that fails a lot and matters a lot when it fails, that might be a great place to try a pilot. If it is truely a problem area, and given you and the team you work with can make a significant difference in how it’s performing, it’s got to receive serious attention. Once you’ve achieved that, the natural question asked is “what made such a difference”? There lies your opportunity to show how your team made this happen. The fact that it was a team effort, proven successful and not too hard to accomplish has got to start getting the attention of others in your organization. What better way to get a program to take hold then to show success and value for all to see. 

Give it a try, it just might make a ‘big’ difference.

If we perform corrective work too soon, we use resources unnecessarily. We also risk destabilizing a stable system that can result in an unplanned equipment failure.

If we wait too long to perform corrective work, repair costs will be higher because substantial secondary damage will occur. This damage can cause problems in other areas such as productivity, quality, safety and environment.

To avoid these costly scenarios, we need to understand what the actual state our equipment is in and provide the appropriate maintenance at the right time.
Inspection results are often vague and subjective because they tend not to accurately specify the severity of equipment condition.

Once we have enough data about equipment that we have inspected using these severity states, we can track and trend equipment performance. This will provide the information we need to perform the right corrective work at the right time.

Next week: Visit our blog to read tips on defining failure and consistency!

Tips for Establishing Effective Key Performance Indicators

by Joe Marin, Implementation Consultant, Ivara Corporation

The old saying that goes “What gets measured gets done” is only partly correct.

Many organizations find that despite a seaming wealth of Key Performance Indicators (KPI’s), little gain can be seen in the bottom-line of the corporation or at least tied defensibly with the initial project investment to achieve those gains. How then is a widely accepted process management tool failing to provide what it is meant to?

The answer is that performance management is not a simple process. Sure, over the long-term, well established processes and performance measures require minimal effort to maintain when compared with managing out-of-control processes, but to get from the out-of-control process to the “well-oiled” version takes a concerted and well reasoned effort. Key performance indicators can help in this regard but care must be taken in their application. This is meant as an introduction to many of the aspects to consider when establishing and using key performance indicators to measure improvement.

DETERMINE BUSINESS NEED
First determine what is important to the business, make sure that the indicator is linked to the business’ objectives (ie, Safety, Environmental compliance, Cost effectiveness, Quality, etc). There are dozens of potential business objectives, select only the most important to start with, those that are clearly defined as company policy are obvious choices. This is in keeping with the Six Sigma philosophy of focusing the measures on what is Critical to Satisfaction (CTS) of the customer.

DETERMINE THE AREA OF FOCUS AND LIMIT SCOPE
Walk the process and determine which areas show opportunity for improvement. List the processes and sub-processes in order of the size of opportunity and potential impact to the business objectives. Select one to three of the sub-processes for improvement. Note that, the more processes you try to affect change on at the same time, the harder it is to determine which changes you made to the process are responsible for the changes measured with the KPI’s.

Limit the amount of business objectives you intend to measure the impact on to one or two. Over time, the amount of measures will grow and cover all business objectives eventually, however, it is important to always watch your scope. It is a common failing to establish a large amount of performance measures all at the same time; this shot-gun approach overwhelms those who monitor and take action, and greatly reduces the chances that the implementation will be successful.

4-I’S – INDICATORS INEFFECTIVE IN ISOLATION
A single indicator will only tell part of the story. An indicator can only tell you what the output of the process it is measuring is. Only by comparing the indicator with other related measures can the problem be refined. For instance, Schedule Compliance = 40%, what does this tell us as a business? It tells us that we have an issue with schedule compliance, but it does not say why there is a problem or where the source is. Combine this with Estimate Accuracy = 90%, Maintenance Labour Utilization = 115% and %Reactive Work = 75%, and we can more closely theorize the root cause of the problem. Just looking at the Scheduling Compliance measure, the business can often be mistaken in its interpretation of the measure and assume that because the problem appears to be with scheduling, the solution also lies within scheduling. Whereas, when we look at the related data, Estimate accuracy, Maintenance Labour Utilization and % Reactive Work, it appears more likely that a large amount of reactive work likely forced the maintenance group to break the schedule.

KPIS SHOULD BE DESIGNED AND TESTED
Indicators generally return numeric values, which leads to the assumption that the results are irrefutable, scientific and that the conclusions based on the results are objective and not subjective. Most indicators can be interpreted subjectively. In order to aid the results to be as objective as possible, a KPI set should be formulated carefully by experienced personnel working together; this group should theorize likely reasons that a KPI can return a positive or negative value or trend and document this to aid those interpreting the data. Each theory should be tested with positive and negative inputs of the actual object/sub-process being measured prior to implementing the KPIs.
Do not just establish KPIs that measure areas thought to be performing poorly. Although we want to focus on making improvements, it is important to maintain what is already performing well.

It is often a good idea to include personnel that are well educated in Six Sigma methodology to aid in the development of KPIs, as they are knowledgeable in scientific measures and can link KPI sets to the Critical to Satisfaction measures of the business objectives.

LEADING VS LAGGING INDICATORS
Leading indicators are those that are used to predict early changes in performance and are direct measurements of a specific step in a process (hence the term leading, they lead you to the specific source of the measurement before the cycle is complete). Used effectively, leading indicators can be used to mitigate specific problems within a process before the business is impacted negatively.

Lagging indicators are those that show trends or the effects of changes to a set of steps within a process (hence the term lagging, they measure the results of a process after the cycle is complete). Lagging indicators are good gauges of cultural or systemic problems within a process. Used effectively, lagging indicators can be used to show the results of changes made to specific actions taken to improve the process.

If you’re trying to making improvements to a process, use leading indicators to determine those areas you will want to make specific improvements to then use lagging indicators to show the results of the changes made and how they affect the larger (parent) process. Leading and lagging indicators should be used in concert with each other, not in isolation.

IMPLEMENT IN STAGES
When implementing KPI’s, ensure that they are implemented in stages. There are two main reasons for this. The first reason, as mentioned previously, is to not overwhelm those monitoring and acting on the indicators. This can lead to sustainability issues with the measures and missed opportunities for improvement. The second reason is to allow time for the KPI’s to be tested and refined. When first starting to develop performance indicators it is common to have minor issues with the data inputs and calculations that lead to inaccurate measures. Having too many indicators often buries these flaws and often leads to wasted effort. This sows the seeds of doubt in the reliability of all key performance indicators and threatens the sustainability of the program.

In keeping with the 4-I’s, key performance indicators should be implemented in related sets (modules) in these implementation stages.

ESTABLISH THE BUSINESS TIERS IN THE REPORTING STRUCTURE
Indicators should role up the business tiers from the front-line managers to upper management. That way, upper management can determine if the performance issues are systemic or local.

SET APPROPRIATE SAMPLING FREQUENCY AND TREND WINDOW
Because most key performance indicators are used to aid in determining process effectiveness, the indicators must have time to build effective trends in order to determine if the process being measured is improving, maintaining or degrading. Ensure that you determine an effective sampling frequency and trend window for each individual indicator. Do not use the performance results to take action or report gains based on only one or two readings, always allow the indicator to develop a firm trend, highlight any outliers in the data and record the reasons for each outlier as this helps to defend the data’s integrity.

UNDERSTAND HUMAN FACTORS
Always be aware of the human factors involved when establishing KPI’s. Workers can often feel threatened that their own performance is what is being measured and can find ways of manipulating the process or measure to ensure a favourable measure results. Once a measure is compromised it can take a fair amount of effort to restore it of develop a new measure to take its place. So how do we deal with the human factors?

1. Educate all personnel in the reporting chain of the purpose and philosophy of the KPI.
2. Do not use a KPI as a personal performance measure or blame tool. KPI’s measure a process’s effectiveness and are not intended to measure the poor performance of a worker. There may be reasons that a worker’s performance will affect a process negatively, i.e. lack of training, oversight, etc. The moment that a KPI is used to directly reward or punish a worker, the KPI will likely be compromised to ensure that only a favourable output is returned in the future.
3. When establishing a KPI set, do not focus on solely the negatives. Processes that are working need to be reinforced, therefore processes that show upward trends consistently need to well publicized, but also, processes that are maintaining themselves at or above targets need to be promoted because these processes are being effectively executed by personnel and providing good value to the business.
4. KPIs should be automated and require as little direct effort to input values whenever possible. This not only helps the KPI to be sustainable, it has the added value of better protecting the integrity of the data and does not have the negative stigma of being an onerous chore for those inputting data.

IN THE END
Like all new tools, KPIs can be very useful in helping a business to achieve and maintain improved performance but, like all new tools, adequate time, development, learning and careful use is needed to ensure they produce what is intended of them.