Choosing the measurement method that fits the job

Why it matters

The success of a measurement is judged not by how much data it collects but by whether it answers the decision question. Knowing how long a defined work cycle of a part takes and knowing what share of a worker's time is spent waiting are different goals. The first needs a start and an end; the second needs activity classes at suitably chosen moments. If a tool is chosen without making this distinction, the needed quantity cannot be calculated even when a tidy table is obtained. In short repetitive work, continuous timing can follow the work cycles and their elements. In long, scattered work, sampling can give information about the activity share without the burden of watching the whole time without a break. Choosing a method starts with this question.

The frequency of repetition, access, visibility and how rare an event is all affect feasibility. A short job with clear cycle boundaries is suited to direct time recording. In work that is spread across long preparations and different tasks, a shared work cycle definition may be weak; the activity share may be the more meaningful target. But sampling does not guarantee catching the start and end of a rare event. Even if the share of waiting can be estimated, the duration of a single event stays open. For this reason the two methods are not used interchangeably at random. Sometimes first the prevalence of the loss is investigated by sampling, and then the time structure in the selected division of work is investigated by continuous timing. The question of each stage is written separately.

Video can be useful for re-examining the boundaries of the observed elements. With Yamazo Studio, a video time study can be made by marking the elements on the video. This capability does not make an invisible boundary visible, and it does not correct a wrong sample scope. Reviewing the same recording again does not create a new independent work cycle. Likewise, taking short recordings for work sampling may be easy; but going only to the easy hours spoils representativeness. The question of the job and the measurable events come before the features of the tool. The image, time and recording conditions of the tool to be used are assessed afterwards, according to the requirements of the chosen method.

The time budget is also not just how long the work takes. The observer prepares, gains access, processes the record and checks it. Calendar time and effective observer time are different quantities. A plan spread over several days can have a short effective load; a short recording may need a detailed review. The budget calculation must cover all the necessary work. But fitting within the budget does not make a method scientifically adequate. If there is a missing-data, dependence or visibility problem, the plan is corrected. If the suitable method cannot be carried out within the current budget, the question is narrowed or the resource requirement is explained; a different quantity is not measured and presented as if the real question had been answered.

Formula