Manufacturing – Qsutra

What can we Discover from the Process Data by Creating a Simple Histogram?

Umesh — Mon, 05 Oct 2020 12:39:59 +0000

A simple demonstration of Histogram usage in your process!

Histogram is commonly used for graphical representation of numerical data. Most importantly, it is one of the basic quality improvement tools used in various industrial processes.

What is a Histogram?

A Histogram is a graphical representation of data where data values are shown in rectangular bars which shows the occurrence (frequency) of data values on numerical intervals of equal size.

By plotting histogram, we can examine the shape and spread of our process data. I would like to highlight the terms “shape” and “spread”. The shape of a histogram can help us to determine whether the data is normal or not. While the spread of histogram can help us to determine variability in data.

Now, let’s demonstrate one simple example – How can we use the histogram in manufacturing process scenarios?

Suppose in a Battery Manufacturing Industry, a quality control engineer needs to ensure that the process is producing the plates within the target value. The target plate weight value is 20 kg. Since plate dimension is an important factor in the battery manufacturing process.

The engineer collects a sample of 75 plates by random sampling and measures the weights of it. From the measured data, a simple histogram is created.

Now, my question for you – “Whether this process is producing a plate within the target weight value or not?” So, what will be our interpretation….

Here, a greater number of plates were produced with a plate weight of 14 to 24. But our target value is 20. Moreover, many plates were beyond the target. Therefore, this process is bad and have to work for improvement.

I guess this brief example will help you to understand the usage of a histogram in your process.

Benefits

Some important considerations

NB – Want to know more about Histogram and Applications in Industries? Attend our Statistical Training Program starting from basic to advanced level. Some of the Statistical training certified courses are Predictive Analytics Masterclass, Essential Statistics For Business Analytics, SPC Masterclass, DOE Masterclass, etc. Apart from Statistical training, we also conduct Minitab Certified Training Program, starting from basic to advanced level. Some of the Minitab software training certified courses are Minitab Essentials, Statistical Tools for Pharmaceuticals, Statistical Quality Analysis & Factorial Designs, etc.

Is your Sample Size enough for doing the Analysis?

Umesh — Mon, 06 Jul 2020 05:39:46 +0000

I guess this is an important question we should ask ourselves before doing any kind of analysis. We should know the sample size required to do the analysis. Say, if we collect more samples than required – what will happen? We will waste our resources and time. Thus enhanced cost and complexity. On the other hand, if we collect fewer samples than our test may fail and hence less accuracy on results. On an extreme note, our test may fail to detect the important effect and could not meet the test objective. Hence we should determine the optimum sample size required for analysis.

Let’s discuss a very basic example to understand “why the determination of sample size is important in data analysis?” Generally, salt is a very important factor to determine the taste of food. We all know how much work is needed during the cooking process, right! Now, the hard work will be paid off if only we get a significant result i.e. food must be tasty. Say, if we put more salt than required or less salt than required – what will happen? We all know the consequence, right!

To avoid such circumstances in data analysis, “Concept of Power and sample size” comes into play.

What is a Power and Sample size?

It is a concept of estimating adequate sample size required for data analysis to achieve significant results. We already know sample size – it is a count of individuals or samples required for analysis. But you might be thinking about the term “power”. Power analysis tells us a number of sample size required to avoid any type of error in decision making. In other words, we can say that it is a probability of making the right decision irrespective of errors.

I would like to mention a few factors which affect the power analysis are sample size, variability and alpha value (level of significance). We should choose adequate samples so it will not hamper the power analysis. There could be a variation (spreadness of a data) in a process and thus results in a random error. We should use proper sampling procedure and measurement analysis to minimize variation in a process. Alpha value represents the probability of making the wrong decision when it is actually true. The common alpha values are 1% (0.01) and 5% (0.05).

In a ball bearing manufacturing industry, engineers are concerned that the bearing diameter has shifted away from the target of 0.6 cm. They consider a difference of 0.01 cm to warrant adjusting the equipment. From the historical data, the standard deviation is 0.005 cm.

Now the question is “How much data is needed for analysis?”

With reference from the power curve – for doing the analysis we need a sample size of 5 where alpha value is 0.05 and power is 0.9.

What are the common power values used?

When determining the sample size, we should put the optimum power value as to avoid any kind of error. Suppose in a test, if it has low power value than it might fail to detect an effect and if it has a higher power value than small effects might seem to be significant. Some of the common power values used are 0.8 and 0.9. Let’s say if your study has 90% power than it has a 90% chance of detecting an effect that exists.

There are various ways to determine power and sample size. The easiest and simplest way to determine it is by using the “Power and Sample size” option in MINITAB. With Recent Release, Minitab now Available on both Desktop and the Cloud.

NB – Want to learn, How to run Power & Sample Size Analysis in Minitab Software? Attend our Minitab Certified Training Program, starting from basic to advanced level. Some of the Minitab software training certified courses are Minitab Essentials, Statistical Tools for Pharmaceuticals, Statistical Quality Analysis & Factorial Designs, etc. Apart from Minitab training, we also conduct basic and advanced Statistical training. Some of the Statistical training certified courses are Predictive Analytics Masterclass, Essential Statistics For Business Analytics, SPC Masterclass, DOE Masterclass, etc.

Also we offer solutions to build a robust Enterprise Decisions Management system that can be very effective for organisational Decision Making.

How to Decide whether to Accept or Reject the lot in Manufacturing Scenarios?

Umesh — Sun, 07 Jun 2020 12:42:57 +0000

Suppose in an organisation, when we have to evaluate the quality of incoming or outgoing products – “How we do it?” Based on certain quality characteristics – “How we make a decision whether to accept or reject the lot?”

The answer lies behind the concept of “Acceptance sampling”. It is an alternative to 100% inspection. Here, inspection is performed on a lot or batch of parts or products. We can use acceptance sampling to accept or reject a lot based on a representative sample. It plays a major role in the quality control process.

Please keep in note that acceptance sampling doesn’t improve the quality of a process but it helps to control quality to a great extent. It helps in decision making whether to accept or reject lots.

Some of the benefits are

Basic Terms

Lot size – It is the entire set of items/observations from where samples are drawn.
Sample size – It is the number of items that are drawn randomly from a lot for inspection.
Acceptance number – It is the maximum number of defects or defectives that are allowed in an acceptable lot.
Acceptable Quality Level (AQL) – It is the highest proportion of defectives in a lot or batch that is considered acceptable. It describes “What the sampling plan will accept?”
Rejectable Quality Level (RQL) – It is the highest defective rate that the consumer is willing to tolerate in a lot. Lots at this level are considered to be poor and have a low probability of acceptance. It describes “What the sampling plan will reject?”

Which method to use: Attribute sampling plan or Variable sampling plan?

It depends on which types of data we are dealing with. If it is an attribute type of data, we will go for Attribute sampling plan and if it is a variable data, we will go for Variable sampling plan.

Attribute sampling plan – This sampling plan is used to accept or reject a lot based on the number of defects or defectives (discrete data) in a sample. For example, you receive a lot of 5000 capacitors and by this method, you will be able to determine the sample size (how many capacitors you need to inspect) and acceptance number (how many defects are allowed within the sample).
Variable sampling plan – This sampling plan is used to accept or reject a lot based on variable data (continuous data) such as length, thickness, diameter, hardness, etc. For example, you receive a lot of 3000 pistons and you need to verify that the diameter meets specifications. By this method, you will be able to determine the sample size (how many pistons you need to inspect) and criteria for accepting or rejecting a lot (critical distance).

Since we have discussed the basic concepts of acceptance sampling. Now, I would like to move a little forward and discuss the risks involved in acceptance sampling.

You must be thinking, “Why risks are involved?” Let me remind you, we are not doing 100% inspection – we are taking a sample from a lot for inspection.

There are two types of risks involved while doing it –

NB – Want to learn, How to run Acceptance Sampling Analysis in Minitab Software? Attend our Minitab Certified Training Program, starting from basic to advanced level. Some of the Minitab software training certified courses are Minitab Essentials, Statistical Tools for Pharmaceuticals, Statistical Quality Analysis & Factorial Designs, etc. Apart from Minitab training, we also conduct basic and advanced Statistical training. Some of the Statistical training certified courses are Predictive Analytics Masterclass, Essential Statistics For Business Analytics, SPC Masterclass, DOE Masterclass, etc.

We also provide a wide range of Business Analytics Solutions and Business Consulting Services for Organisations to make data-driven decisions and thus enhance their decision support systems.

Understand the Application of ANOVA in Manufacturing Process!

Umesh — Wed, 08 Apr 2020 13:06:36 +0000

Suppose in the Manufacturing Process, we want to compare and check which are the most reliable procedures, materials, etc. We can use the ANOVA test to compare different suppliers and select the best available.

ANOVA (Analysis of Variance) is used when we have more than two sample groups and determine whether there are any statistically significant differences between the means of two or more independent sample groups. In other words, we can say that it checks the impact of one or more factors by comparing the means of different samples. When we have less than or equal to two sample groups – we go for one sample and two sample tests.

What are the types of ANOVA?

There are two main types of ANOVA viz. one–way ANOVA and two–way ANOVA.

One Way ANOVA – It is also known as one factor ANOVA. Here, we are using one criterion variable (or called as a factor) and analyze the difference between more than two sample groups. Suppose in glass industry, we want to compare the variation of three batches (glass) for their average weight (factor).
Two Way ANOVA – Here, we are using two independent variables (factors) and analyze the difference between more than two sample groups. Similarly, we want to compare the variation of three batches of glass w.r.t weight and hardness (two factors).

We have discussed the basic concepts of ANOVA when we have one factor, we use one-way ANOVA and when we have two factors, we use two-way ANOVA. I guess this concept is clear and understandable.

Now, let’s discuss some of the assumptions which we should keep in mind while performing ANOVA.

Example

One Factor ANOVA – In an automobile industry, three quality inspectors (A, B, C) measure the breaking strength of car seat fabric and the management wants to test for a difference between their measurements by comparing means.

Reference

Null Hypothesis – It is a type of hypothesis which is presumed to be true i.e. no statistical significance exists in a set of given observations.
Alternative Hypothesis – It is alternative or contradictory to the null hypothesis. There is a significant difference in the process.
Level of significance (α) – It represents the probability of making the wrong decision when the null hypothesis is true.
p-value – It is used in hypothesis testing for making a decision whether to accept or reject the null hypothesis.

Since our null hypothesis is rejected, we need to find out which sample group is different and so we can work for improvement.

Let’s make it clear that

Post-ANOVA Analysis

We will use the Tukey test (also known as Honestly significant difference – HSD test). With reference to the above example for one way ANOVA, our null hypothesis was rejected and we will conduct Tukey test by using MINITAB.

Grouping Information Using the Tukey Method and 95% Confidence

Means that do not share a letter are significantly different.

NB – Want to learn the concepts of ANOVA and Application in Industries? Attend our Statistical Training Program starting from basic to advanced level. Some of the Statistical training certified courses are Predictive Analytics Masterclass, Essential Statistics For Business Analytics, SPC Masterclass, DOE Masterclass, etc. Apart from Statistical training, we also conduct Minitab Certified Training Program, starting from basic to advanced level. Some of the Minitab software training certified courses are Minitab Essentials, Statistical Tools for Pharmaceuticals, Statistical Quality Analysis & Factorial Designs, etc.

We also provide a wide range of Business Analytics Solutions and Business Consulting Services for Organisations to make data-driven decisions and thus enhance their decision support systems.

Is your Data Reliable enough to do the Analysis?

Umesh — Mon, 09 Mar 2020 05:23:03 +0000

Data – What we need to do the analysis? But how much we are sure that the data we have collected for analysis is reliable enough! The main agenda of this post is to discuss data reliability, it’s importance and how to achieve it.

Let’s start with a brief example, suppose during a Clinical Trial “If the data are not taken properly?” And proceed further for analysis. Then “What will be the consequences of it?” Clearly, it may fail later and results in loss of lives, resources and time. Thus, the device measuring the host condition should produce reliable results.

Now, I would like to introduce the term called “Measurement System Analysis (MSA)” which helps to determine the reliability of a data for analysis. You must be curious to know about it, right!

Generally, MSA is a method to determine the measurement system in a process. The measurement system is a collection of measurement devices, measurement procedures and operators that are used to obtain a measurement. MSA helps us to detect the variation (i.e difference or disparity) exists in a measurement system. So we should evaluate our measurement system before conducting statistical process control, design of experiment, Predictive Analytics, or other statistical analysis. By doing so, we can be sure that our data are reliable for analysis.

We can apply MSA for two types of data namely attribute and continuous data.

Attribute data - When data collected are in the form of count or categorized in the form of group. For e.g. accept or reject, good or bad, and so on. In such scenario, we use Attribute agreement analysis.
Continuous data - When data collected are in the form of fractional or decimal value. For e.g. weight, length, time, temperature, etc. In such scenario, we use a Gage R & R analysis.

MSA is classified into Attribute agreement analysis and Gage R&R analysis based on data types used.

Example

Suppose in a Wheel Manufacturing Company, operators use gage to measure the diameter of the wheel. Are we 100% sure that the collected data is measured correctly or not?

With reference from the Gage Run Chart – Three operators measure 10 wheels, three times per wheel randomly. Here, we can examine the differences in measurements between operators. We can see that Operator B does not measure consistently, and Operator C usually measures lower than the other operators.

Let’s discuss a few terminologies used in measurement system analysis.

Accuracy – It is the difference between the measurement of a part and its actual value.
Bias – It is the measurement difference between the observed value and true value.
Linearity – It is the difference or a change in bias value w.r.t its size.
Stability – It is a measurement of how well the system performs over time or we can say, it is a change in bias value over time.
Precision – It is the closeness between the measured value and the true value.
Repeatability – It is a variation which occurs when the same operator measures the same part with the same device.
Reproducibility – It is a variation which occurs when the different operators measure the same part with the same device.

Why MSA is important?

It is a prerequisite for successfully implementing Lean Manufacturing, Six Sigma and Lean Six Sigma projects. It is also useful in the implementation of SPC, DOE, Enterprise Prediction Management, Enterprise Risk Management, Enterprise Risk Management, and Enterprise Quality Management processes. It gives the better reliability of the measurement system in a statistical way.

NB – Want to learn, How to run Measurement System Analysis in Minitab Software? Attend our Minitab Certified Training Program, starting from basic to advanced level. Some of the Minitab software training certified courses are Minitab Essentials, Statistical Tools for Pharmaceuticals, Statistical Quality Analysis & Factorial Designs, etc. Apart from Minitab training, we also conduct basic and advanced Statistical training. Some of the Statistical training certified courses are Predictive Analytics Masterclass, Essential Statistics For Business Analytics, SPC Masterclass, DOE Masterclass, etc.

We provide a wide range of Analytics & AI Solutions like Business Analytics, Digital Process Automation, Enterprise Information Management, Enterprise Decisions Management and Business Consulting Services for Organisations to enhance their decision support systems.

How Statistics Applications can add value to the Industrial Manufacturing?

Umesh — Sat, 08 Feb 2020 07:24:29 +0000

A brief illustration of the Applications of Statistical Methods in the Manufacturing Industry.

Statistics, we heard a lot about it right! In this data-centric world, the importance of statistics is well known in all the areas. It holds a key position in general manufacturing, healthcare, finance, pharmaceutical, automotive and so forth. With the expertise of modern Statistical Data Analytics Software, we can do statistical data computations in seconds.

The main agenda of this blog is to illustrate its application in Industrial Manufacturing. Manufacturing has emerged as one of the highest growth sectors in terms of market size, investments and coverage. It comprises of apparel, chemical, automobiles, electronics, aeronautical, electrical, textile, food, etc.

By applying statistics, we can present data in a simpler form and make a data-driven decision. There are many business benefits of statistics, such as process efficiency & productivity, better decision support systems, quality excellence, predict your business outcomes, and many more. Moreover, statistics is the base level knowledge required in all the Data Science disciplines (Business Analytics, Predictive Analytics, Machine Learning).

Now, let’s discuss a few of the popular statistical methods used in the Manufacturing Industry.

We often do Sampling in a process, now you must be thinking “How sampling is related to hypothesis testing?” To draw inference or conclusion about the sample data w.r.t the population as a whole, we use hypothesis testing. There are different methods used in hypothesis testing based on sample, such as one-sample t-test, two-sample t-test, ANOVA, etc. To know more, you may visit our page “Hypothesis Testing’’.

Suppose in a battery manufacturing industry, a quality inspector wants to compare the performance of batteries manufactured in different shifts. So, what he will do? He will collect a sample from different shifts and perform Hypothesis Testing. Here he will use the concept of ANOVA (Analysis of Variance).

To read more about ANOVA, you may check our blog titled “Understand the Applications of ANOVA in Manufacturing Process!”.

There are four different shifts – Shift 1, Shift 2, Shift 3 and Shift 4. Since our p-value is 0.004 (From ANOVA Table) which is less than our alpha value 0.05, we can conclude that the means are significantly different from each other. Similarly, on an interval plot, we can see the mean differences between each shift.

We collect data for analysis and draw some conclusion from it. Now the question is “How much we are sure that the data we collected is reliable enough for analysis?” MSA helps us to find variation in the data we collected from a measurement system and guide us to improve the system for measurement. Measurement System Analysis has different methods based on the types of data, for continuous data – Gage R&R Analysis and for attribute data – Attribute Agreement Analysis.

Suppose in a piston manufacturing company, operators use callipers to measure the diameter of the piston. Now in this scenario, “Are we 100% sure that the collected data is measured correctly or not?” To overcome it, we will use Measurement System Analysis. Thus, it is always the first priority to evaluate measurement system before conducting any kind of analysis. We will create a simple Gage Run Chart to examine the differences in their measurement.

Three operators measure 10 pistons, three times per piston randomly. Here, we can examine the differences in measurements between operators. We can see that Operator B does not measure consistently, and Operator C usually measures lower than the other operators.

We want to know whether our process is statistically under control or out of control. And further, we want to evaluate whether our process is capable to meet our client requirement or not. How to examine it? Here, Statistical Process Control (SPC) comes into play which is a tool through which we can control, monitor and improve the process. It is used to track variation and eliminate it from a process. It is widely used in almost all manufacturing processes for process stability and continuous improvements.

Suppose a glass manufacturer wants to determine whether the production process for a new product is in control. They collect subgroups of 10 units every hour for 10 hours and assess the thickness of glass by plotting in a Control Chart.

One point is out-of-control on the X-bar chart. We can conclude that the process is not in control and should be improved. We can do brainstorming to find the root cause of a problem.

The terms like robustness and optimization are at peak in manufacturing processes. How to achieve it? Simple, by conducting the design of experiment or popularly known as DOE. It provides a quick and effective method for optimizing products or processes. It plays a major role in new product design & formulation, process development and improvement.

Suppose in a simple manufacturing process – we need input variables (raw materials) which ultimately change into the product at the end of a process. There are various noise factors which could affect the process such as temperature, changes in raw material, part variations, power fluctuation, etc. To get the robust process and optimize response value, we apply DOE on it.

After creating, data collection and analysing the DOE Design, we have created a Surface Plot to understand visually. To achieve good quality, the optimal settings should be between 400° and 450° and duration between approximately 30 and 36 minutes.

We have discussed a few of the statistical methods which are popularly used in the manufacturing industry. This small post can’t cover all the statistical methods used in the industry. But at least, we got a glimpse of it.

Statistics has emerged as one of the basic foundational skill needed in data science and applications. Most of the advanced statistical methods are used for quality improvement and organisational excellence.

NB – Want to learn the concepts of Statistics and Application in Industries? Attend our Statistical Training Program starting from basic to advanced level. Some of the Statistical training certified courses are Predictive Analytics Masterclass, Essential Statistics For Business Analytics, SPC Masterclass, DOE Masterclass, etc. Apart from Statistical training, we also conduct Minitab Certified Training Program, starting from basic to advanced level. Some of the Minitab software training certified courses are Minitab Essentials, Statistical Tools for Pharmaceuticals, Statistical Quality Analysis & Factorial Designs, etc.

We also provide a wide range of Business Analytics Solutions and Business Consulting Services for Organisations to make data-driven decisions and thus enhance their decision support systems.