IAF Methods

Histogram

by for . Last edit was over 2 years ago
30 - 120 any

Creating a graphic that demonstrates visually how raw data is useful

0
0
Share
Embed
Use
Edit
View
Delete
DRAFT Pending Declined
Submit for approval
Decline
Approve

Additional info

Goal

The intent of this method is to monitor the performance of a system, to discover and track variations in certain processes and to distinguish between special and common causes of variation.

Attachments

You will be able to upload attachments once after you create the method.

Materials

  • There are no additional resources needed.

Instructions

Before

Setting: A computer in order to create the chart is recommended, however the chart can also be hand drawn.

Number of participants: The group can be of any size that can work together. 

Time needed: The time needed strongly depends on the extent of the data, and how precise the analysis is wanted.

Ideal conditions: There are no special conditions needed.

Pre-Work Required: Gathering the data that will be graphed.

Facilitator personality fit: The facilitator does not need any particular characteristics, but should be familiar with a computer, and a program that might help to create the graphic.

During

The first step is to find, from the raw numbers (data), the highest and lowest value. This will be the range for what follows.
Next, the numbers of bars that will be used have to be determined. Here, attention has to be paid to the number of bars, as too much would mean that the pattern could become lost in the details. On the other hand, when using to few bars, the pattern maybe lost within the bars.
By
dividing the range by the number of bars, the width of each bar can be determined. Starting with the lowest value, define the grouping of values to be contained or represented by each bar.
A compilation table, as displayed in the attachment, has to be created and the boundaries for each grouping have to be filled in.
Then, fill the compilation table by counting the quantity of data points for each bar, and then the total amount of data points in each bar.
After counting the data, the horizontal and vertical axes can be drawn and labeled.
Now, the bars can be drawn into the graphic so as to correspond with the totals from the frequency table.
As
a last step, the patterns of variation can be identified and classified according to the pictures in the attachment and to the brief description below.

Bell Shaped: The normal pattern

Double Peaked: Suggests two distributions

Skewed: Look for other processes in the tail Truncated: Look for reasons for sharp end of distribution or pattern

Ragged Plateau: No single clear process or pattern

After

Follow-Up Required: The client will have to analyze the collected data further so as to improve possible problems, and if applicable take the necessary actions.

Usual or Expected Outcomes: The outcome should be a histogram, and hence, an analysis for the client with which he/she can work further so as to improve certain factors in the organization or company.

Potential pitfalls: Unavailable

How success is evaluated: The process is successful when a clear pattern (type of histogram) could be created with the data, so as to give the client a clear "picture" of the processes, with which can be worked with further.

Background

Source: http://www.qaproject.org/methods/reshistorgram.html

Recognizable components: An important step of this process is the monitoring of work(ers).

References: An important step of this process is the monitoring of work(ers).

Comments (1) ( 3.0  avg / 1 ratings)

Please Log in or Register for FREE SessionLab account to be able to comment or rate.
  • The product is useful: the process for creating a histogram with a group could be spelled out more clearly.

    Delete
    over 2 years ago