Select Page
[Total: 0    Average: 0/5]
Based on a statistical analysis of frequencies, this method allows to determine operations times by freeing of technical difficulty or judgement.

## Introduction

It is a technique by quantitative analysis to measure in terms of time of human activities, machines or any other observable activities.

Developed in the years 1930 by L. H. C. Tippett in the textile industry, this method is based on a statistical study of frequencies to determine the standard time. It is especially used to analyze non-repetitive or irregular activities.

It consists of observing the work at random times over time and analyzing the proportion of the observations corresponding to each of the activities studied. It then deducts the time of the operation.

Less accurate than other techniques, it has the advantage of being simple, fast and inexpensive.

## Procedure

### 1. Choose the tasks to be analyzed

You have to cut the work into the task that you want to analyze. For example, work can be separated into tasks such as assembly, transport,… or measure the distribution of ” administrative ” Tasks ( batch files…) among operational tasks.

### 2. Take a first measure

From the data we want to analyze, we need to get a first idea to prepare the recording. For this, a first series of measurement is carried out in a random way. As a result of this, a numerical estimate of the different parameters that one wishes to analyse is carried out.

For example, following a first analysis, the ” machine ” time is identified as 20% of the total cycle time.

### 3. Analysis with the study by sampling

The size of the sample needed to obtain a reliable statistical result is calculated. It is obvious that at the most there are measures, at the most the results will be reliable. However, for cost and time issues, the number of measurements is determined to be the most accurate in relation to the desired accuracy. There are two techniques.

#### The statistical formula

Statistically, the formula for calculating the Sample size Depends on the size of the mother population and different factors.

#### The Graphical method

Another method is a graphical analysis called Nomogram. Source: D. P. Adams (1964)-Nomography: Theory and application

### 4. Prepare the observation

This statistical method is based on a ” random ” survey of data. We will plan the measures to ensure that this condition is met. For this, two possibilities:

• a fixed interval: For example, every 10 minutes, we will carry out the reading.
• random“: There are many possibilities such as pulling a ” pile or face ” to find out if you are reading at the moment.

It is also at this stage that we will build the survey sheet. It contains:

• An introduction with the date and the observer.
• The number of readings to be carried out.
• The various tasks to be tackled. ### 5. Observing and raising data

The survey is carried out according to the random rule chosen. Each time he has to make a statement, he observes the current task and notes a bar in the box of the task he observes at the moment.

### 6. Analyze the data and conclude

The method is to calculate the proportion of events (observed stains) on a given sequence. Of this proportion, we deduct the standard time of the event calculated according to the following formula:

Standard time of one unit = (TObs * TCar * JP) / NbPts + Increase

With:

• TObs: Total time of observation in minute.
• TCar: Proportion of the characteristic in%
• JP: Value of Pace judgment
• NbPtr: Total number of parts produced

For example, we observed 100 times 1mn of operation. On these 100 times we produced 20 pieces and we observed 16 times the drilling operation, 6 times a Recovery, 14 times a move to the stock and 8 times the assembly operation. Whereas we do not have a surcharge and the JA is 100%, we deduce:

• Drilling operation: 109 secs per Part
• Covers: 41 secs per Part
• Moving to Stock: 95 secs per Part
• The mounting operation: 54 seconds per Part

## Source

G. Salvendy (2001) – Handbook of Industrial engineering: Technology and Operations management

O. D. Salih, A. Raouf, J. D. Campbell (1999) – Planning and control of maintenance systems: Modeling and analysis

J. P. Tanner (1991) – Manufacturing Engineering

L. H. C. Tippett (1935) – Statistical Methods in textile research: a snap reading method of making time studies of machines and operatives in factory surveys

D. F. Sittig (1993) – Work Sampling: A statistical approach to evaluation of the effect of computers on Work patterns in the healthcare industry