a. Selecting the job

Reasons behind the selection of job for time study are

• i. If the job is new.
• ii. Change in the method of existing time standard.
• iii. Complaint from worker or unions regarding time standard.
• iv. For bottleneck operation which delays the consecutive activities
• v. Change in the management policy for time standards i.e. wage incentive plans.

Thumb rules for selecting job

• i. Bottleneck operations
• ii. Repetitive jobs
• iii. More laborious jobs
• iv. Longer cycle time of job
• v. Frequent overtime in department or section

b. Selection of worker

• i. Selected worker should be skilled with good temperament.
• ii. Pace performance of selected worker should be closed to average so the observed times are near to the normal times
• iii. For group of workers - do the time study on more than one qualified worker.
• iv. Differentiate between a qualified worker and representative worker
• v. Representative worker – skill & performance is average.
• vi. Qualified worker – ILO defines, One who is accepted as having the necessary physical attributes, who possesses the required intelligence and education, who has acquired necessary skill and knowledge to carry out the work in hand to satisfactory standards of safety, quantity & quality.

c. Breaking the job into elements

After collection of information regarding the job, the next step is breaking the job into elements.

Element: - A distinct part of a specified job selected for observation, measurement and analysis.

Work cycle: - Complete sequence of elements necessary to perform specified activity or job.

Reasons for breaking the jobs into elements

• i. To differentiate productive time from unproductive activities.
• ii. To access the rate of performance more accurately.
• iii. To identify the different types of element providing approximate treatment to each element.
• iv. To isolate high degree of fatigue elements for providing more accurate fatigue allowances
• v. To distinguish machine elements from human elements.
• vi. For accurate rating.

d. Number of cycles to be timed

Number of cycles to be observed is directly proportional to the degree of accuracy desired. The number of observations at 95% confidence level & accuracy of $\pm$ 5% is provided by following formula.

$\mathbf{\eta}=\left[40 \frac{\sqrt{n^{\prime} \sum x^{2}-\left(\sum x^{2}\right)}}{\sum x}\right]^{2}$

Where,

$n^{\prime}=$ Number of preliminary readings

$n=$ sample size (number of observations)

$\sum x=$ sum of preliminary set of observations.