Workers may dread being hauled in for annual performance reviews, but distribution center managers just can't get enough feedback—at least when it comes to the DC operation as a whole. A recent survey by the Warehousing Education and Research Council (WERC) indicated that respondents had a strong desire to establish more performance metrics for their DC operations (though only 37 percent of those surveyed said they used labor standards). And the survey respondents were surprisingly eager to compare their own performance to that of other companies: a full 89 percent expressed a willingness to participate in benchmarking.
That's certainly a step in the right direction, but it's all too easy to get off on the wrong track when developing a benchmarking program. Some companies focus on the wrong targets. Benchmarking based on such metrics as on-time performance, order fill rates, and order cycle time in and of itself will not go too far in improving performance.
Others choose the wrong kinds of benchmarking partners. Many times, comparing performance with external parties tells you little more than which companies are performing better than yours is … or are just as confused as you are. And there's always the danger that managers of a company that's outperforming the others will become complacent.
Whatever external benchmarking programs you participate in, don't let them divert your attention from internal benchmarking. Benchmarking internal operations against past performance is critical to the success of any DC operation.Without some record of where you've been, there's no roadmap for where you're going. The DC manager must keep in mind, however, that if internal measurements yield excellent results, it's possible that the operation is achieving outstanding performance at an unnecessarily high cost.
Since approximately 80 percent of DC dollars spent today go for labor, the real management challenge lies in controlling labor costs. Measurements like cases or unit loads per person-hour and orders or order lines per person-hour will yield valuable information on both absolute costs and productivity trends. Before these can be measured effectively, however, there must be a standard, preferably an engineered standard, for the work being performed.
Essentially, a "standard time" is the amount of time it should take an experienced worker to complete a specific task in a recurring environment. For example, you could develop a standard time for moving a pallet with a fork truck from a dock to a bay location. Comparing actual performance against this standard will indicate how efficiently the forklift operator is working.
There are several acceptable methods of establishing standards, among them estimations based on experience and work sampling. But the most accurate will be a formal time study, which can be as simple as having an analyst observe and, using a stopwatch, time one worker over a number of cycles. This standard is then applied to all workers performing the same task. (It's important that the observer thoroughly understand the tasks because the more "creative" worker may add an extra motion or two while under observation to ensure a lower, more relaxed standard.) Keep in mind that all but the shortest tasks should be broken down into basic segments so the standard won't be distorted by a task (stretch wrapping, for example) that may not be performed every time.
Once standards have been established for all DC functions and management is convinced that the work force is performing up to standard, then meaningful metrics can be developed and benchmarked. At this point, the DC manager will be able to measure employee performance with the assurance that the base from which the measurements begin is as good as it can be.
Gartner, Inc.
