The first practical step to implementing your 7% solution is to tackle the largest segment of non-value added time. That is, the 0.94 hours lost when average performance is below ‘standard’
This is a typical representation of individual performances. I have used actual WMS transactional data for a picking department taken from a past, real life, case study where the distribution pattern is represented as a ‘normal distribution’ (the bell curve).
The range of performances shows the mean, or average, performance around the 80% mark, with individuals on either side split half and half, either achieving performance below or above the average, with a wide spread from low to high.
The data seems reasonable, at first glance, but there are two main points to address for a business that is attempting the ‘7% solution’ challenge.
1. The first point is that the average performance level is 80%. Although this doesn’t necessarily ring alarm bells it does indicate that there is room for improvement. The performance calculation for this data set was based on a production target of 230 picks per hour. 80% performance means that this department is actually achieving 184 picks per hour.
That means that it takes on average 1.25 hours to pick 230 items instead of the target of 1 hour. This represents the ‘lost’ time relating to the effect of a lower ‘measured performance’ illustrated in the pie chart segment from part 1. You can multiply the extra quarter of an hour by the number of shift hours and by the number of pickers on the department to calculate the total ‘lost’ time.
The 7% solution TIP #1
Take a look at the distribution of individual pick rates using your WMS data but don’t use the average as your benchmark. Look at the top end of the league table for your best performers. You will see a much wider gap if you look at the difference between top and bottom, and this is more likely to get you asking questions. Talk to your best pickers about how they organise themselves and manage their workload. Now you can benchmark your other, lower performing pickers and start to look at how you can develop performance improvement programmes. Next time you look at the distribution graph, you want to see the average moving to the right, say from 80% to 85%.
2. The second point is to question whether 230 units per hour is the correct target or ‘standard’ to be using. In this case study, when a work measurement (or time study) exercise was completed the Standard Minute Value (SMV) for this pick task came out at 0.16, or 375 units per hour. This value was validated by two Industrial Engineers taking observations across multiple shifts at different times of the day.
The new performance distribution, for the same WMS data set, looked like this;
The average performance is now around 50%, but we can still see that the performance range includes pickers who are picking at a rate of 80% and above, against the increased ‘standard’ target of 375 per hour.
It is not unusual for large adjustments in the SMV to be identified where a business takes the first steps towards moving from production targets based on historical information to validation of methods and times using formal measurement techniques.
The 7% solution TIP #2
I cannot stress enough the value of using experts and specialists to validate your KPIs. At face value, in this case study, it would seem reasonable to take a sample of WMS transactional data and to be convinced that a target of 230 units per hour that gave an average performance level of 80% was entirely reasonable.
However, if you are relying on your WMS data alone to provide you with your production targets then you will have another opportunity to deliver on your ‘7% solution’ challenge. You may have to invest in some specialist resources for a short period of time, but as you can see, it will prove its value very quickly.
You wouldn’t want to make such a dramatic change to your targets in one leap, but once you have established the correct standards you can start to make managed adjustments to your KPIs.