Measuring productivity is an important facet of turnaround management. The objective is to reduce waste through a more efficient utilization of the available resources.

The following steps should be taken in order to be able to measure productivity in an objective manner:

• A detailed work order scope must be defined, estimated and reviewed.
• A good, workable schedule must be prepared, from which manpower staffing charts can be prepared.
• Daily "Force Reports" reflecting the actual number of workers present at the site should be submitted to the planner.
• Worker’s time cards (or time sheets) should be coded with the correct Work Order numbers and should list the type of work to be performed.
• Daily updating and preparation of progress reports must include all crafts and contractors involved.

The above provides all of the basic information required to evaluate productivity (earned value analysis). The planner can then compare the actual productive manhours (from the time cards or time sheet summary) against the "earned manhours" calculated on the daily progress report. If more manhours are used than earned, it could indicate either poor productivity or estimates which are too low.

To determine which is the case: whether productivity is high or low, or if the estimates are high or low, it is necessary to determine if the manhours are high or low across the entire estimate, or only in some specific work orders. If the manhours are consistently high (or low) across the entire estimate, then the estimate might have to be revised. If only a few work orders are involved, then they should be investigated to see whether any activities or tasks were either overlooked or underestimated. Or perhaps extra work arising from inspections could boost the amount of work for which no estimate was foreseen in the original scope.

Another method that can be used to determine the accuracy of the time and manpower estimates is through sampling: selecting a few representative work orders, and "clocking" specific tasks as they are being performed.

Sometimes estimates can be off due to a change in the approach to performing the work. For example, bundle pulling could have been defined as a labor intensive operation, involving a winch truck, a cherry picker, and several craftsmen, resulting in a high manhour estimate. Instead, a bundle puller machine could be used, operated by one or two men. This could make a substantial difference in the total estimate, depending on the number of heat exchangers involved. Similarly, coke deposit removal in a vessel could have been originally scoped as chemical cleaning, but due to the extent and hardness of the actual sediment, it could involve several men with chipping or gouging tools, plus laborers to haul off the waste.

It is obvious now from the above, that any other methods used for determining productivity (especially the "interview" type) are totally inadequate and can result in grossly misleading conclusions.

Keeping excellent records over a period of time will allow for improving the estimates for future turnarounds. A good estimate will, of course, give a more objective and realistic baseline to measure productivity against.