Notes

Strengths and weaknesses

Strengths:

• Short study duration – Kaizen Events occur over periods of 3-7 days
• Implements real changes/improvements – the Kaizen Event takes process mapping and redesign to the next level by rapidly implementing the process revisions, which creates instant results (i.e. instant process time savings)
• High acceptance levels – because there are representatives from all participant groups in the Kaizen team, acceptance levels of the Kaizen solutions tend to be generally quite high (as the changes are made to suit everyone’s needs, not just the needs of some). It also helps to have a champion from each of the participant groups to help teach the process changes to the rest of his/her colleagues.
• Well rounded analysis – the Kaizen Event results are gathered through both collaborative discussion and empirical evidence

Weaknesses:

• High resource use – Kaizen Events require the full time participation of several staff members throughout the course of the event
• Short-term observational results can be misleading –only short periods of time are spent empirically observing the process workflow, which could lead to unrepresentative results
• The Hawthorne effect – it can be argued that the observation of subjects causes them to change their behaviour; thus, causing the observation results to be unrepresentative

Return to tabs

Plan

Sample selection

Sample selection

The sample selection phase for a Work-Sample Study is almost the same as the sample selection phase of Time-Motion studies (see Sample selection under the Plan section of Time-Motion studies), except for the fAct that because the Work-Sample observations are instant, a much larger sample subject base can be selected. Unlike Time-Motion studies, where the emphasis is on performing detailed observations of few subjects, Work-Sample studies aim to perform several instant observation samples on many subjects; thus, promoting a greater Study breadth, by covering a larger population.

Return to tabs

Category development

Category development

The ‘Category development’ stage for a Work-Sample Study is the same as in a Time-Motion Study. Please see Category development under the Plan section of Time-Motion studies for details.

Return to tabs

Select observation schedule

Select observation schedule

Scheduling the observations for Work-Sample studies can be a much more difficult task than scheduling observations for Time-Motion studies. The first challenge with scheduling Work-Sample observations is determining how many observations are needed to estimate the process details. Generally speaking, the more precise the approximation needs to be, or the more tasks are in a given process, the more observations need to be made. For example, (Brisley, 1971) states that for a task that occupied 50% of a worker’s time, for a 10% precision level, 400 observations would have to be made, and for a 1% precision level, 40,000 observations would need to be made. To further clarify this jump in required observations for increased Study precision or for capturing frequently changing tasks, (Brisley, 1971) also states that for a 1% precision level on a task that occupied 5% of a worker’s time, 630,000 observations would have to be made, and for a task that occupied 1% of a worker’s time, 4 million observations would have to be made. Two methods can be used to determine this, the statistical method and the ‘Nomogram’ method. These methods are described, in detail in (Matias, 2001).

The second challenge in setting a Work-Sample observation schedule is determining when to conduct the observation samples. There are three forms of observation scheduling for Work-Sample studies, fixed, variable, and self-reported logging.

Return to tabs

Fixed scheduling

Fixed scheduling

Fixed scheduling is when a set schedule of repeat observation times is used for the observation Plan. For example, a fixed schedule would be to conduct the observations seven days a week, 4 times a day, at 12:00am, 6:00am, 12:00pm, and 6:00pm. Fixed scheduling is advantageous in that it creates an easy-to-follow guide for the observers, as to when they need to make their observations.

The Downside to fixed scheduling is that it creates a strong possibility of introducing systemic biases into the Study. Systemic biases occur when routine measures capture routine Activities. For example, consider the following routine, hourly schedule for physician at a local clinic:

• from on the hour to quarter past the hour, the physicians perform their administrative functions;
• from quarter-past the hour to five-to the hour, the physicians tend to their patients;
• and from five-to the hour to the hour, the physicians take their coffee breaks.

If routine scheduling was used in this example to determine the work prActices of the physicians, then depending on the number of observations, and the timing of the observations, there is a strong chance that the Study would form an unrepresentative view of the physician’s work prActices. For example, if the fixed Work-Sample schedule was set to every hour, on the hour, then all the observers would see the physicians Doing would be performing administrative functions or having coffee breaks. Because Work-Sample studies use these workflow ‘snapshots’ to estimate the entire process workflow, in this example, the observers would estimate that the physicians spend all of their time performing administrative functions and having coffee breaks (because their set observational schedule would never observe the physicians tending to their patients).

One way to counterAct systemic biases in a fixed schedule Work-Sample Study is to schedule more observations. The more observation samples are made in a Work-Sample Study, the closer the complete set of samples will approximate the complete process. For instance, in the above mentioned case of capturing the physicians’ Activities at their local prActice, if an observation was made every twenty minutes, as opposed to every hour, the observers would notice that the observations made within the hour had a much higher likelihood of the physicians performing direct patient care than the observations made on the hour. Another method used to reduce systemic bias is variable scheduling.

Return to tabs

Variable scheduling

Variable scheduling

Variable scheduling is where ranDom times are set for the observation schedule (i.e. one day the observation schedule could be 12:30am, 6:00pm, and 8:17pm, while the next day could be 4:20am, 12:00pm, 3:30pm, 6:45pm, etc.). The ranDomness of the variable schedule lowers the risk of systemic bias, as it breaks away from routine measure capturing routine tasks. For instance, in the case of measuring the physicians’ work prActices, it would make no difference if the physicians routinely performed their administrative duties and took their coffee breaks on the hour, because the ranDomness of the observation schedule would mean that the observations would be made throughout the hour, capturing a mix of the physicians’ routine Activities (i.e. a patient visit at 8:17am, another patient visit at 11:48am, a coffee break at 12:57pm, etc.). Variable schedules are, however, much more difficult to Plan for in prActice, than fixed schedules, because the observers Don’t have a routine schedule to follow. Variable schedules can also make resource Planning difficult, since the observers may have long waits in-between observations at one point in time, and no breaks in-between observations at other times.

Return to tabs

Self reported logs

Self reported logs

In the alternate form of Work-Sample studies (where the Study subjects maintain journals of their work-process Activities), scheduling becomes less of a problem for the Study designers, and more of a problem for the subjects, as they have to take time off of their normal work to write Down their Activities. This is normally performed on an ad-hoc basis (whenever the subjects have a free minute to jot Down their Activities), but it should be encouraged that the subjects record their Activities as frequently as possible, to prevent problems with lapses in memory.

Return to tabs

Selecting the observers

Selecting the observers

The ‘Selecting the observers’ stage for an observed Work-Sample Study is the same as in a Time-Motion Study. Please see Selecting the observers under the Plan section of Time-Motion studies for details. For a log-based Work-Sample Study, no observers need to be selected.

Return to tabs

Software and hardware selection

Software and hardware selection

The ‘Software and hardware selection’ stage for both an observed and a log-based Work-Sample Study is the same as in a Time-Motion Study. In an observed Work-Sample Study, the observers will require the process recording equipment. For a log-based Work-Sample Study, the subjects would benefit from the use of the equipment. Please see Software and hardware selection under the Plan section of Time-Motion studies for details.

Return to tabs

Training observers

Training observers

The ‘Training observers’ stage for a Work-Sample Study is the same as in a Time-Motion Study. Please see Training observers under the Plan section of Time-Motion studies for details. For a self-reported process log Work-Sample Study, no observers need to be selected.

Return to tabs

Do

Data collection

During the data collection phase of Work-Sample studies, the observers will travel from subject to subject at the Planned observation scheduling periods, and take time-stamped snapshots of the current process tasks that the subjects are performing. Much like a Time-Motion Study, the process tasks are recorded in the data collection tool using codes from the process task coding scheme, the name of the observer is recorded (to indicate who observed the task being performed), the name of the subject is recorded (to indicate who performed the task), and additional notes may be recorded (to help describe the task if future task clarification is needed). Unlike a Time-Motion Study, however, the duration of the observed tasks are not recorded in a Work-Sample Study, as Work-Sample studies estimate the timing of tasks from samples, as opposed to measuring the timing directly. When the observations are complete, they are loaded into a computer for future analysis and secure storage.

In the second form of Work-Sample studies, the data collection process can be achieved through gathering self-maintained subject process logs. For this method, the Study subjects are asked to maintain journals of their process Activities, including data about what was performed and when. This form of Work-Sample Study Does not require observers; however, it places an additional burden on the subjects (since they have to observe themselves). This form of data collection can potentially create self-reporting biases (where the subjects record that they did one thing, and really did another).

Survey

The ‘Survey’ stage for a Work-Sample Study is the same as in a Time-Motion Study. Please see Survey under the Do section of Time-Motion studies for details.

Return to tabs

Study

The Study stage for a Work-Sample Study is the same as in a Time-Motion Study.

Please see Study under the Time-Motion section for details.

Return to tabs

Act

The Act stage for a Work-Sample Study is the same as in a Time-Motion Study.

Please see Act under the Time-Motion section for details.

Return to tabs

PDSA Summary

Work-Sample studies involve an observer taking ‘snap-shot’ observations of subjects performing process tasks in the field at either variable or fixed intervals and then combining these snap-shots to create an estimate of the whole process. Alternatively, subjects may keep task logs where they record what they were Doing at what times and researchers then combine these task logs.

Plan

• Select the Study sample based on the user profiles – Which user groups will be monitored?
• Develop the Study measurement coding scheme – What will be monitored?
• Create the observation schedule – When will the observations take place?
• Select the observers – Who will collect the Study data?
• Obtain needed software and hardware – What tools are needed to collect the data?
• Train the observers – How will the observers collect the Study data?

Do

• Collect the observational data
• Survey the users to clarify the observational data

Study

• Use the data to develop a well-defined map of the current work process

Act

• Either summarize the Study results into a Time-Motion evaluative process report, or use the results in a process refinement methoDology (e.g., Kaizen Event)

Return to tabs

References

References

• Brisley, C. L. (1971). Work Sampling. In H. B. Maynard (Ed.), Industrial Engineering Handbook. New York: McGraw Hill.
• Matias, A. C. (2001). Work Measurement: Principles and Techniques. In G. Salvendy (Ed.), Handbook of Industrial Engineering (3rd Edition ed., pp. 1448-1457). Institute of Industrial Engineers.

Return to tabs