Diagram or control chart: process control tool
I present another of the considered the 7 tools of Quality. The Diagram or control graph serves as a process control tool. This tool is widely used, and it helps us to evaluate in a visual way, the stability of the process, as well as to identify those values ​​that emerge from the system’s confidence range.

In this article I explain in detail what is the graph or control diagram? How is it What aspect do you have? What are the control charts for? How can I make these graphs? What sources of variation do we have in a process? What types of control charts exist? What does the control diagram give us to our ISO 9001: 2015 quality management system? And to finish a simple example, in Excel, to control the process.

What is a graphic or control diagram?

 

It was created by Walter Andrew Shewhart in 1920, the chart or control chart, also known as Shewhart diagram, control chart or process behavior diagram, is one of the 7 quality tools defined by Ishikawa.

 

A control chart is a diagram that is used to examine whether a process is in a stable condition, or to ensure that it is maintained in that condition.

It allows us to know, in a graphic way, if all the points in our sample are within a certain range.

In statistics, it is said that a process is stable (or is in control) when the only causes of variation present are those of a random type. For these cases, inferences can be made regarding the output of the process, that is, the quality characteristic that is being measured. Instead, the presence of special or assignable causes causes the process to destabilize, preventing the prediction of its future behavior.

Based on the information obtained at certain time intervals, the control charts define a confidence interval: If a process is stable, 95% of the time the result will remain within that range.

How is it What aspect do you have?

The structure of the graphs contains:

– A “central line” (LC) that represents the average values
– A top line that marks the “upper limit of control” (LCS), which is considered the most average
– A lower line that marks the “lower limit of control” (LCI) that represents the mean less
– Finally, the points points that contain information on the readings are represented; They can be average of groups of readings, or their ranks, or the individual readings themselves.

The control limits mark the interval of confidence in which the points are expected to fall.

The chart looks like this:

 

What are the control charts for?

The control charts serve to:

– Determine the state of control of a process.
– Diagnose the behavior of a process in time.
– Indicate whether a process has improved or has worsened.
– It allows to identify the two sources of variation of a process
– It serves as a tool for detecting problems.

How can I make these graphs?

Usually, in companies where the use of control charts is widespread, statistical software is adapted to the needs of the company. There are numerous computer programs in the market that allow you to perform this type of graphics, among others, in an automated way.

But, it is also common to find people who perform their control charts in the Excel software.

What sources of variation do we have in a process?

The sources of Variation in a Process are attributable to the attributable causes or the common causes:

1. Assignable or Special Causes

These are the sporadic factors that destabilize the system. Your identification is immediate and easy.

2. Common or Natural Causes

They are the factors that little affect the variability of the system.
Its presence is random and not easily detectable.
They are usually related to administrative aspects.

How do we begin to make a control chart?

Before establishing a control chart, you need to clearly define the following points:

1) the objective of the graph: what do we do about this graphic?
2) the aspect to be considered: what do we want to control, what process we must control, what characteristic to be measured, perhaps length, perhaps number of defects, the volume …
3) the unit from where the sample is to be taken: where the sample is to be taken.
4) Determine what type of graphic should be done

5) Determines the time in which you will be capturing the data and defining with baseline

in the type of graphic that you are going to trace, questions such as the quantity of samples to consider (consider at least 20) and the size of each.6) Compile the data7) Calculate the average values ​​and the range (Upper limit E inferior) 8) Represents the graph9) Analyze the result of the data obtained. What types of control charts exist? There are numerous criteria for performing a control diagram, in this article I will show you, the most common way, but you must apply The one that best controls your process.

The most common way is to calculate the average, and the limits of calculate considering 3 times the standard deviation,

you will see it in the example that I will put to you later. We can have control graphs of variables or control of attributes: 1. Variable Control Charts: The quality characteristic that is measured is a continuous variable (Mass, Length,Temperature, Force …). If that is the case, we can find graphs based on the average () and the range.- Graph – R Averages and ranges- Graph – s Averages and Standard Deviation- Graph – R Mediums and Ranges- Graphics – R Individual Readings and Ranges2 . Attribute Control Charts: Think of an attribute-based quality feature such as compliance with a requirement. We do it with discrete variables. From this, four types of graphics are derived: – Graph p Percentage of defective units or processes – Graph np Number of defective units or processes – Graph c Number of defects per opportunity area – Graph u Percentage of defects per opportunity area Control graphs They are commonly used to monitor the statistical control of the process or SPC for its acronym in English.

This latest tool, as well as the APQP, PPAP, AMEF and MS, belong to the group of tools known as the Core Tools in the automotive sector, which is a requirement of ISO / TS 16949. Example in Excel Our example is a company that manufactures elements Prefabricated concrete For example, we are going to control the defects detected in the manufacture of prefabricated alveolar slabs. In this case, we will not distinguish by the type of detected defect. We will note, by day of manufacture, the number of defects found. The results can be seen in the following table: * The data collected in this table are invented, by way of example. Once we have the table above, the first quadrant, where we have collected all the data we want to analyze, we must Determine the middle line and the lower and upper limits. Next I explain how to calculate it: To do this, we must first calculate the mean and the standard deviation: To calculate the mean defects per day, it is done by dividing the number of defects detected between the total number of manufacturing days: With all the previous data , We must set the lower and upper limits. Usually the company, or the person who wants to control the process, is the one who decides the criterion to apply. The most common thing is that they are set as the average more / minus twice the standard deviation, since you ensure that 95.4% of the data enter within the range.

The percentages would be:

The mean +/- 1 time the standard deviation (68.3% of the data)

The average +/- 2 times the standard deviation (95.4% of the data)

the average +/- 3 times the standard deviation (99.7 % Of the data)

In our case we have considered the standard deviation twice. With all the previous data, we will obtain the table with the following columns to prepare the graph: Once we have the table above, we select the entire range of data, included The titles, we insert a chart of lines in Excel,

and we already have the control chart, we will only have to tweak the design, adding the title, adjusting colors and so on. The graph with the previous data would be as follows: In this case, we would have all the data within the range of 95%, the only value that would be out of range, is the last. It would be necessary to analyze why.

What gives the control chart to our ISO 9001: 2015 quality management system?

One of the requirements of the ISO 9001: 2015 standard is the control of the processes, as well as the analysis Of the results of each one of them. Through this tool we can control and analyze graphically the different processes of the company. When we talk about control diagram, we think only that it can be for the control of production processes or measurement of some magnitude, but it can also be used for any other process. I will give you an example, for the human resources department: A company, which has 200 workers, wants to control, in a graphic way, the training of these workers. The number of training courses taught by the company to train their workers, for each one, is counted. The average is established, and its range is determined

Diagram or control chart: process control tool

Uso de cookies

Este sitio web utiliza cookies para que usted tenga la mejor experiencia de usuario. Si continúa navegando está dando su consentimiento para la aceptación de las mencionadas cookies y la aceptación de nuestra política de cookies, pinche el enlace para mayor información.

ACEPTAR
Aviso de cookies