Technique 2.56 Matrix Chart (M2)
{product-noshow 9|name|cart|picture|link|border|menuid:206|pricedis3|pricetax1}
The Matrix Chart is a holistic/systematic approach to understanding one's own operation and how it compares with others. It involves understanding how changing one measure affects the measurement of others. It was developed in order to explain the relationships that exist between measures and to identify what drives change. This is essential to understanding the success of your organisation.
In order to identify those measures that are critical to your organisation, there is a need to look at the entire operation as a whole by
. Identifying the boundaries, inputs and outputs of the operation and developing a flow chart of the task and sub-processes; this is linked with an understanding of the groups, teams and individuals who play an important part in the organisation.
. Choosing the appropriate measures (NB: a leading indicator will provide information that is in anticipation of an action, while a lagging indicator will provide information on the results of an action)
. Grouping the measures, eg performance, cost, quality and safety, etc. Each of these large indicators will contain sub-measures. Training is a contributor to performance; and may include personal development with the outcome an increase in productivity. It is vital to collect accurate, suitable, consistent data/information on each measure, and realise the links between different measures. There are generally 3 types of measures, ie
i. the smallest value, eg costs
ii. the largest value, eg productivity
iii. a target value, eg training
It is important to identify what data type is appropriate for each measure.
. Both management and workers are involved
In evaluating the whole picture, the following questions need to be answered
. What does the performance data really mean?
. What are the interrelationships of the measures?
. Has one measure improved at the expense of other measure(s)?
One of the main problems plaguing any form of data presentation is that people find it difficult to extract meaningful information from tables of data. Instead of focusing on the data, we should pay more attention to the practices that drive performance as shown in the matrix chart analysis.
Designing a Matrix Chart
The matrix chart consists of a set of concentric circles ranging from the inner-most circle having a value of 1.0 to the outer-most circle which has a value of 0. The chart is broken up by spokes radiating from the center, with each spoke representing a measure. The target measure is 1.0. The better an operational performance is, the closer the results data point gets to the center of the chart (see template)
Each ratio must be set up so that all resultant values always fall between 1.0 and 0.0. Though the data is normalised, the magnitude of the gap between the target and the comparable operation is always preserved
How to calculate and interpret the points
Largest value = operational data divided by benchmark
Smallest value = benchmark divided by operational data
Target value = benchmark - (benchmark - operational data) divided by benchmark
Each data point for an operation is calculated and the resultant points are plotted on their respective spokes in the charts. Once all the points are plotted, they are connected to each other by a solid line. What develops is a profile of how a specific operation relates to the measures used to describe the operation. Usually the profile is then shaded to indicate the relative area within the points. Better performance is indicated by less shaded area. The data points that are furthest from the center identify the major improvement opportunities.
Below are 2 examples. The first is from Toyota
The above chart shows that the organisation has good training numbers while the other indicators are average. Based on this, there is a need to understand how training is focused and why the benefits are not seen. The training may be focused on an area that has not been recorded as critical. Thus the need for training needs to be evaluated. The chart can be used to track trends within an organisation over time.
(sources: James Madigan, 1992; Toyota, 2006)