Work sampling analysis is a method used to measure the time required for an employee to complete a unit of work. It may be used with scientific studies, quality improvement programs, and workplace-related training programs. It can track how long each worker takes for each task. The effective use of sampling helps an organization solve problems and improve productivity by minimizing idle time, reducing costs, determining staffing levels, and speeding up the production process. Here are six keys to implementing an effective sampling procedure:
- Determine what you want to measure – This includes defining the purpose of measuring, deciding on what you will be measuring, and determining how it will be done. Also, determine if manual or automated measurement will be used.
- Decide on your timeframe for measuring – This refers to deciding how often you will measure. It is important to remember that the frequency (time interval between measurements) must be consistent for schedule control and time standards. Also, changes in the process need to be detected when they occur. A popular approach is referred to as the five-times rule. Take a sample five consecutive times each business day (e.g., 9 am, 10 am, 11 am, 12 pm, and 1 pm). Track any increase or decrease in time required over these intervals separately.
- Identify work activity categories – This includes deciding which activities should be measured. For example, are all activities evaluated equally? Should there be an emphasis on some activities over other activities? It would be important to consider how individual tasks are related to each other because this will help determine if it is appropriate to combine or separately measure them. One good method for developing a list of representative tasks is an affinity diagram that allows group members to input ideas and later prioritizes them by voting with dots (one per idea).
- Choose representative activities among selected categories- This is the most important step and involves choosing random samples from your elected list of representative tasks. These samples must represent the entire range of task time expected in the process. The best way is to select samples from the longest task times first, followed by progressively shorter task times.
- Select random samples from representative activities – This refers to selecting a sample of activities within each category selected in Step 3. For example, if you evaluate ten tasks, choose one activity for each at regular intervals throughout the day or week (e.g., 1 pm, 2 pm). If you compare two processes that together include 100 tasks, then take ten samples at regular intervals (i.e., once every two hours) throughout each process and collect data on all 100 tasks twice during the workday (once for each process).
- Prepare a sampling datasheet – This refers to creating a work sampling data sheet with time estimates listed across the top and down the side. The left-hand column allows for notes that may be important when analyzing or discussing results, while all other columns are used to record time intervals spent on tasks during each interval sampled. Make sure you have many more time intervals than there are tasks sampled so that they can be arranged in order of length (i.e., longest first). If necessary, use decimals to reduce empty space. Also, make sure your layout can accommodate additional rows if more data are subsequently collected.
Work sampling is a valuable way to ensure there is always enough staff on hand. It also helps businesses save time and money because the workload can be monitored more easily without relying on subjective data.
