Data Analysis and Customer Service

The strong link between customer service and data analysis has started to become quite apparent in recent years. It’s no longer about collecting simple metrics like call duration and other similar points. We now have the capacity to analyze customer service interactions on a much deeper, more complex level, and the benefits of that are starting to manifest themselves in many areas of the industry. And we’re likely only seeing the tip of the iceberg, as well.

Integrating Data Collection

You have to start by integrating data collection systems into your current operation. This can be done in multiple different ways, depending on the exact specific details of your operations. When it comes to customer service, there are multiple viable points for collecting data which can then be analyzed in detail. And it’s a good idea to dig as deeply as possible, ensuring that you always have as much as you can of the bigger picture.

Analyzing Efficiently

You also have to make sure that the analysis you’re performing is actually efficient and up to date with your current requirements. Sometimes you might spend a lot of effort on something that will ultimately prove fruitless. Directing your efforts in analysis is going to be very important when you want to ensure that your customer service stays at a top-level at all times. It’s true that our analytical power is increasing over time, but this doesn’t mean that it’s okay to use it inefficiently. You’ll have to direct your approach very carefully in fact.

Revising Your Situation Based on Findings

Sometimes you might come to the conclusion that it’s necessary to make some changes to your current situation based on findings that you’ve made in your analysis. And in some cases, the changes necessary are going to be very minor. But in other situations, you’ll need to completely throw some of your established ideas out of the window. This can take a lot of time and effort, and you should be confident in the quality of your findings if you want to ensure that you’re not wasting your resources on something that will ultimately prove fruitless. When your actions are backed by solid data as we described above, this will be much easier.

Expanding Your Collection Practices

Data can be collected in many different ways, and it’s a good idea to study the range of available options in detail if you want to be sure that you’re not missing any points that could be relevant to you. Don’t just focus on one or two analytical positions try to study as much as you can about your current situation, and hook those analytical systems up to different places to see what kinds of results you’re going to get. If you notice that one of your collection systems is not producing any useful information that you’re applying on a regular basis, it might be a good idea to consider cutting down on your expenses in that area.

Complex Analytical Systems

As the complexity of your data collection grows, so should your analytical systems as well. It’s important that you’re able to process all of the data you’re collecting adequately, instead of just piling it on. Otherwise, you’re only going to make things more difficult for yourself in the future, when you find yourself working with large data sets that are practically impossible to wrangle int heir state.

As long as you pay attention to the data that truly matters to the operation of your business, you should be able to improve your customer service significantly by just collecting and analyzing as much information as you can about the way things are currently running. And before you know it, you’ll be a market leader in terms of customer service and satisfaction.

The post Data Analysis and Customer Service appeared first on 6sigma.

Lean Six Sigma (LSS) is becoming more and more established in various industries and sectors, and it’s slowly but steadily taking over the market. It’s clear that it has applications even outside fields like production and development, and there is a particular interest in the use of Lean Six Sigma in marketing recently that’s been growing steadier and steadier. Indeed, it does look like the methodologies found in LSS can be of great use to those trying to develop a successful marketing campaign.

Integrate Data Collection

Having access to enough data to make adequate decisions on can be a huge boost to your productivity, with or without using Lean Six Sigma on top of that. Simply integrating some advanced data collection systems into your practices can make a significant change to the way you’re able to process your input and make decisions in the future, and that alone can result in some of the best increases in productivity you can realize. Keep in mind that your data collection needs to be organized appropriately for this to work though, as you don’t want to simply gather everything indiscriminately that’s a fast way to ensure that you’re going to get buried in irrelevant data sets.

Trim the Waste

On that note, Lean Six Sigma dictates that it’s very important to always ensure that you’re reducing the waste in your operations by whatever means necessary. Marketing can suffer particularly badly from waste piling on too quickly, and if you’re not careful about reducing your data sets appropriately and sanitizing your input, you could find yourself in a lot of trouble.

On the other hand, when used correctly, LSS can be a fantastic way to ensure that your overall project runs without too much waste and that you remain focused on the points that truly matter.

Iterate Over Previous Versions

Don’t stay stuck on the same version of your marketing campaign for too long make sure that you actually iterate on it based on what you’ve learned from previous experiments. LSS can quickly reveal various points about your operations that may not be immediately obvious, but it’s not a magic wand that’s simply going to resolve those issues automatically. It’s up to you to take action and implement what you’ve learned through LSS in order to see long-term success.

This means that you have to be willing to take a closer look at the kinds of mistakes you’ve made in previous projects, and be ready to make the necessary changes to them. Don’t just make those changes randomly though ensure that you’re properly oriented towards improvement, and know what the long-term implications of everything you do are. Adjusting the parameters of your project based on actual knowledge and true facts is the best way to ensure that you’re evolving it in the right direction.

Stay Flexible

This means that you have to develop your project in a flexible manner from the very beginning you can’t simply work with a rigid model and expect it to perform flawlessly over countless iterations and changes. It’s inevitable that you’re going to need to implement some changes in that project eventually even without the use of LSS in the first place but if you don’t have a good, flexible foundation to work with in the first place, that’s not going to happen.

A rigid project can be very difficult to adapt to a more flexible state later on as well, further adding to the difficulty of wrangling a marketing campaign like that and turning it into something useful. Because of this, make sure that you start with something that is open to changes in the first place, and be very careful in how you’re evolving that project at every step to ensure that you don’t compromise its flexibility in the long run.

The post How to Apply Lean Six Sigma to Your Marketing Process appeared first on 6sigma.

Process Execution Management (PEM) is the overall concept of bringing a specific process to completion from start to finish. It’s a broad term that encompasses a wide variety of disciplines and methodologies, and a proper approach to PEM is an important asset in the skillset of any competent leader. You need to have the ability to see things in an organized way, and to plan from early on in order to avoid the most common problems that you can encounter in each area of your operations.

From Planning to Execution

PEM covers both the planning as well as the actual execution of a process, and if you’re just starting to get familiar with the field, the initial planning phase is a good place to focus your attention. You’ll typically be able to resolve a lot of issues without even encountering them directly by just making sure that you have a solid plan of action, and you’ll eventually want to integrate an advanced planning approach into your workflow on some fundamental level.

Of course, it’s impossible to plan every type of process from start to finish, which is why you’ll also need to develop a good ability to change your ideas along the way, and drop plans that might not end up working out like you anticipated in the beginning. This might sound complicated, but it’s just one of the skills that you’re going to pick up along the way if you’re a responsible leader with the right kind of attitude.

An Informed Approach

It’s also important to ensure that you always have every piece of the puzzle when dealing with a situation, and this often means collecting a lot of data and having it available for analysis in a systematic manner. This is a common aspect of lean methodologies nowadays anyway you’ll want to implement various data collection systems in your production chain and in your general workflow, and manage them on a regular basis to ensure that they’re actually capturing the kind of data that you’re interested in.

Then, you’ll also want to make sure that you have a good system for analyzing all of that information that you’re capturing, and this is a highly individual point that different companies are going to approach in different ways. Exploring the different systems available for complex data analysis is something you’ll want to do as early as possible, and you should spare no resources in trying to find the most adequate solution in this area.

Handling Issues Along the Way

It’s inevitable that you’re going to run into some problems while executing your processes, no matter how well you’ve planned everything in the beginning. It’s just the nature of most business processes, and the important thing is that you’re able to deal with those issues in a competent manner and get them out of the way as quickly as possible as they pop up. This is also something that will require you to build up some solid experience in order to truly get the hang of how things work, but rest assured that if you’re dedicated to finding a solution, you should be able to learn a lot along the way.

Note that you may not be able to get rid of absolutely every problem you encounter. Far from that, in fact but as long as you take notes and try to learn something from every situation, that’s all that really matters in the end. No leader is born with all the knowledge they need in order to perform their job properly. It’s something you gather along the way, and it’s important to pay attention to the situations you’re going through in order to truly build up the kind of experience that you need.

Conclusion

A good grasp on Process Execution Management and the best practices when handling different types of issues is a critical asset in the portfolio of any responsible leader, and you’ll want to devote some attention to that as soon as you can find the time in your schedule. This is especially true in cases where you’re directly responsible for overseeing the execution of important processes in your organization, but even if that’s not the case, you can probably expect to get there sooner or later, so it’s best to be prepared.

The post Understanding Process Execution Management (PEM) appeared first on 6sigma.

The use of Six Sigma in the manufacturing environment is well established and has been studied in detail. Many approaches have been developed over time, and today this is a solid field that keeps building on what’s already been done. On the other hand, the transactional environment seems to be understood much more poorly by lean experts, leading to a severe lack of expertise in this area of the market.

There are various challenges in the transactional environment that are unique to it alone. For example, most processes tend to be flexible rather than following a rigid structure, and their flow is altered by various decisions along the way. On the other hand, collecting information is a more difficult task as there are typically fewer reliable sources to gather it from, and it mostly boils down to self-reporting.

It’s clear that an expert working in the transactional environment must have a fundamental understanding of the full range of tools available at their disposal, and understand how each of them can help them perform better in this kind of working environment.

Data Collection and Aggregation

It’s very important to have a good way to collect all the relevant data to the current processes, and to aggregate it in a way that makes it easily accessible to everyone who should have access to it.

Modern technology offers a wide range of tools for data collection, aggregation and retention, and you should look into integrating systems like that in your workflow on a fundamental level. Don’t rely on self-reporting for critical points, this can only lead to problems in the long run if a discrepancy is discovered.

It’s also important to make sure that data is pruned over time, and not just focus on retaining as much of it as possible. One common problem in many organizations operating in the transactional environment is that they eventually get flooded with irrelevant data and find it hard to make a viable decision.

Visualization

Another critical aspect for working in the transactional environment includes the ability to use visualization tools and techniques, and understanding the flows behind them. These are not just useful for high-level leadership, but it can benefit employees at lower ranks within the organization too. When everyone can clearly see how a process flows and what the steps required to complete it are, people tend to make better decisions at each step of those processes.

And this matters because, as we described above, the transactional environment often features a good number of processes which get executed in different ways depending on current conditions, and sometimes this can be quite unpredictable. When everyone has a good way of visualizing what’s going on, decisions can be made in a way that gives predictable results.

Decision-Making Aligned with Current Processes

If you’re going to be effective in a transactional environment, you’ll need to ensure that any decisions taken in the course of executing a process are aligned with the long-term goals of that process. Using tools and enacting processes that emphasize sound decision-making will ultimately help drive forward an organization towards success. This is a point that can often be hard to get across in organizations with old preexisting processes, but it’s important to get it through, because improper decisions can quickly stack up and detract from the current efforts of the organization.

But from the perspective of a low-level employee, it can sometimes make sense to take a decision that ultimately drags the project down. This often comes down to saving time on a personal level, or even across a whole department, and while it’s clearly not beneficial to anyone, it still happens quite often.

Conclusion

The transactional environment presents many unique challenges that are not found in any other working environments. Those who’re already experienced applying lean principles to the manufacturing environment will have to adjust to the many intricate aspects of the field. The good news is that lean is actually very compatible with this style of work, you just need to have the right approach, and know what the best tools available for this purpose are.

The post What Are the Best Tools for Use in a Transactional Environment? appeared first on 6sigma.

Performance measurement is a very effective method for analyzing the way a part of your organization performs or even the whole organization itself. It has been used in many professional circles for quite a while, and it now has an established place in many industrial environments. The methodology has evolved quite a lot since its initial inception, and it’s important to keep some things in mind when applying it — otherwise you risk running into some common problems.

1. Incorrect data collection and analysis methods

Performance measurement relies heavily on good data, so it makes sense that making a mistake in this area can have a very severe negative impact on your results. Nowadays we have many tools at our disposal for the purpose of data collection, retention, and analysis, and it’s important to make good use of everything you have available.

It’s a good idea to have some systems in place to perform sanity checks on the collected data as well, as you never know when one of your collection systems might malfunction in some odd way, producing a set that has a few tiny, but important, details wrong.

Another point to keep in mind is to have a good system in place for pruning irrelevant old data. This can become quite problematic over time, especially if your systems keep changing in specifications every now and then. It’s okay to maintain old data sets that have been captured on a different version of the system, but you have to ensure that there is some process in place to align the different data sets.

2. Not applying the methodology consistently

Performance measurement is not something you should do once and then forget about it. It’s a continuous, ongoing process that must be integrated tightly into your operations if you want to see good results from its application. This means that you should instill this attitude in your employees on every level of the organization, from top to bottom. Make sure that everyone is on board with applying performance measurement, and that there are processes in place to control its use.

This applies especially strongly to the data collection facilities involved in your performance measurement system. Make sure that everyone knows what data points are important in your analysis, and how to organize the collected data to make it more accessible and presentable.

And of course, this also means that whenever your data collection practices change in some way, you must communicate this to everyone involved as clearly as possible. A common problem observed in organizations applying performance measurement is that they communicate changes like these too late or not clearly enough, leading to confusion in the way data is collected. This can subsequently lead to the kinds of problems we mentioned in the first point above, and it’s clearly a situation that should be avoided.

3. Having the wrong idea about good performance

It’s not always easy to define good performance, and many large organizations are still struggling with this on many levels of their work. This is somewhat relevant to the first point in this article, but it’s also a separate issue. If you don’t know what your performance targets are, you might end up moving in the completely wrong direction as a result of the analysis performed on the collected data.

What’s worse, sometimes this kind of mistake can compound over time, requiring you to go back over a significant period to undo the damage. It’s rarely easy to recover from a situation where you’ve been putting a lot of effort in the wrong area of your business, so it’s best to try avoiding this kind of situation altogether in the first place.

On the bright side, once you’ve defined your performance targets properly, and you also have good systems in place for measuring that performance, you can see a great improvement in your operations, and you’ll move towards your target in great strides. At the same time, it will be easy to undo any damage that’s been caused by wrong moves on your side.

The post 3 Common Mistakes in Performance Measurement appeared first on 6sigma.

Check sheets are a tool with a lot of unrealized potential, at least in the way they’re being used in various organizations around the world. While many people see them as nothing more than a checklist that lets you verify that a process has been carried out step by step with perfect accuracy, there is a lot more that a proper check sheet can do for you. Mastering their use as early as possible in your leadership can be a great boost to the productivity of the organization, and learning how to use the different types of check sheets correctly can easily transform the whole way you’re doing your work.

Collecting Data for Future Analysis

By far one of the greatest benefits of a good check sheet is that it can allow you to easily collect a lot of data for statistical analysis. There are several types of check sheets that have been specifically designed for this purpose, actually, and knowing the specific application and benefits of each one can be a tremendous boost to your productivity.

For example, if you’re frequently facing specific issues in your processes and you don’t know where they are coming from, you can use one of the several types of defect-related sheets to get an additional insight into that. Check sheets can be very powerful in determining things like the root cause of a problem, its original source, as well as the frequency at which specific problems tend to manifest themselves within your organization.

The defect type sheet can actually be of great use to companies that frequently deal with certain defects, as they can provide you with a direct link between the occurrences of these problems and other surrounding factors. Every little detail will float to the surface, no matter how irrelevant it might have seemed at first.

So for example, if you’re dealing with an issue that only presents itself at a certain time, it might not be too obvious with regular analytical means. But when you’ve stored all test data in a check sheet, you can start going through it to look for links between specific data points. When you use the right statistical analysis tools to work with the data, those connections should become pretty obvious.

Making Sense of Your Data Collection

A big problem in modern statistical analysis in large companies is that a lot of data is being collected without a clear reason for doing so. Businesses end up investing a lot into infrastructure and collection practices for information that they ultimately may not really need. A check sheet can be great for shedding some light onto the way you’re collecting your information, and keeping the main reason for that collection at the forefront.

The very structure of a check sheet makes it pretty clear why the data within it was being collected in the first place, which is of great benefit to organizations that work with lots of data collection systems. Proper statistical analysis requires you to identify where certain data points came from and why you have them in the first place, and check sheets are one of the best tools for making those systems more rigid and structured.

Of course, using a check sheet alone is no guarantee that your data collection will actually be as sensible as you need it to be, and it’s still up to you to ensure that you’re correctly measuring the different variables that play into this. But it goes a lot towards preventing simple mistakes and allowing you to focus more on the actual data collection instead of working on improving your tools themselves.

Conclusion

Check sheets are tool with multiple powerful applications, and they are one of the best tools around when it comes to working with complex data sets and ensuring that you can perform statistical analysis easily and with no issues. They alone are not enough to ensure that you prevent any issues from occurring in your data collection practices, but they can help you make a lot more sense of the data you’re gathering, and its purpose for your current analysis. Once you’ve properly mastered them, the sky is the limit to how useful they can be.

The post The Connection Between Check Sheets and Data Analysis appeared first on 6sigma.

Determining how well the processes in your organization are running can play a critical role in running things smoothly over an extended period of time. There are many factors that can play a role in the definition of a controlled state, and understanding their full range as well as the relationships between them can be highly beneficial to any serious leader.

The control chart is a tool developed by Walter Shewhart and extended significantly by various others through time which is meant to show whether processes in your organization are under control, and in cases where they are not. The control chart can help you pinpoint the exact sources of deviation.

Control Chart: Basic Concepts

Normally, you would use a control chart to track the performance of tools with a continuous rate of change. However, in some cases it can also be a useful tool for making more binary/discrete comparisons, such as testing against the value of some variable over an extended data set.

The control chart itself is a separate entity from the actual measurement tools used to capture data about the state of each process. You are typically given a lot of freedom in determining how that data is going to be monitored, and you should be familiar with the specific properties of the processes used within your organization.

Once you have your data collection in place, you can think about how you are going to represent it on a control chart. A typical control chart should have some fundamental components, although the exact style is going to vary from one company to another. You are also going to see variance between the different processes in your organization.

You should normally at least have a set of points representing the data you’re measuring. This will look similar to a typical graph plot. However, you are also going to plot the mean of those variables in your control chart. In addition to that, you should also be able to see your specific process limits in the chart.

Using Your Control Charts

When it’s set up correctly, a control chart can quickly show you the relationships between certain process variables and their corresponding appropriate limits. That way, you can tell at a glance if a certain process is under control, that is, if its parameters are within the intended limits. In cases of deviation, you will also see a clear indication of where exactly the problem occurs and how far the data is from expected values.

An important note in the use of control charts is that you are not just looking for deviations that stray too far away from the median. Other results can be interesting too, and a standard rule is to stop production when a certain number of data points in a row all fall on the same side of the average line, e.g. above or below it. You may see the number listed as 7 or 8 in some places, and you are probably going to have your own known optimal values for your organization as well.

Sometimes, certain levels of deviation may even call for a complete stop of your current production. The usefulness of a control chart comes in part from the fact that it can point out some problems in your production before they can be identified by other means.

This tends to work better when you have proper data collection practices in place to track the progress of your facilities over time. As a result, a control chart tends to improve in usefulness, as long as you are tracking everything correctly. As we mentioned above, a successful implementation of a control chart also relies on having a well-developed measurement system in place. The sooner you take a deeper look into that aspect of your organization, the better results you’re going to see.

Conclusion

The control chart is a seemingly simple tool, and it’s true that there aren’t any overly complicated concepts behind it. But once you’ve mastered its use, and you have additionally learned some auxiliary techniques that work in tandem with it, you should see some amazing results in the productivity of your organization and your ability to track the performance of processes.

The post Walter Shewhart and the History of the Control Chart appeared first on 6sigma.

Statistical Process Control (SPC) is a commonly used technique for identifying faults in your production line, and ensuring that the final product is within acceptable quality boundaries. As the name suggests, it relies heavily on statistical methodologies to give you an adequate overview of the current state of your production facilities, and when applied correctly, it can be a very powerful tool for maximizing your output and reducing various kinds of waste. Unsurprisingly, it’s commonly used in lean organizations.

Data Collection

It all starts with gathering all the data that you’ll need in your statistical analysis, and nowadays you have plenty of options for that thanks to modern technology. It’s quite easy to fit your whole production facility with tiny sensors that capture all sorts of important data, and then funnel that into a node that either collects and aggregates the data, or processes it immediately.

Keep in mind that you can go quite far with data collection, and you must always be careful to not overextend your investment in this part of the business. This will not only result in wasted money, but it will also overburden your actual analysis process and make it much more complicated than it needs to be. And that alone can be a huge detriment to the quality of the analysis, therefore it’s crucial to minimize the data collection process as much as your current situation allows you to.

Setting Appropriate Control Limits

Control limits are one of the most important concepts in SPC, and it’s critical that they are set at appropriate levels to minimize incorrect results. This will take a certain amount of experience with your own specific field and the type of product your company makes, and you may also need intricate knowledge of the machines used in the whole process. Sometimes the manufacturers of different production machines may provide you with readily available data for those limits, but more often you’ll have to determine them yourself for your specific use case.

The point of these limits is that no production process is perfect, and there will always be some variation in the output. In many cases though, these variations can be acceptable as they don’t degrade the quality of the final product. Once you’ve set the right limits, you’ll be able to see the important outliers in your production data more easily.

And sometimes, you’ll have to redefine those limits along the way not just when you’ve changed something about the production process, but also when the market itself goes through some changes and forces you to adapt. In some cases this might even mean relaxing the quality control requirements slightly in order to momentarily improve the output capacity of the facility, but care should be taken with this approach to avoid overdoing it.

Reevaluating Your SPC Implementation

Sooner or later you will need to make some changes to the way you’re running your SPC, typically as the company grows and its requirements shift to a new direction. It’s important to regularly reevaluate the way you’re collecting and processing your data, and you should do your best to get your colleagues’ input on this as well. People on other levels of the organization may be able to see certain details that are not as obvious to you, and getting as much feedback as possible on your SPC can be extremely valuable.

Of course, you should also be careful to not overdo this, and if your current analysis produces good results in terms of product quality, then you should focus your efforts on another area of the organization. But never lose focus of the current state of your SPC.

Conclusion

SPC can be a very powerful technique when applied correctly, but it’s not a fire and forget solution. In fact, it’s quite the opposite and can be somewhat demanding in terms of maintenance and attention, but the final results are more than worth it. The impact of a proper SPC implementation on your organization can be incredible, and it’s one of the first steps you should take if you’re having problems with the consistency of your output, or its overall quality.

The post What is the Purpose of Statistical Process Control (SPC)? appeared first on 6sigma.

The Kano model is a great tool for adapting your product/service to the actual demand of your customer base, but it should be used with a degree of caution. It works very well when you already have some experience with the relevant market, but you need some data to drive your decisions, otherwise you won’t get much of a benefit from Kano as opposed to just making random choices.

Understanding Different Needs

The whole basis of the Kano model is splitting the needs of your customers into three categories, and organizing your work so that you can satisfy the needs that really matter in the grand scheme of things. However, this assumes that you can accurately categorize your customers’ needs, as using the wrong category can severely degrade the performance of your organization.

After all, if you believe a feature is of low importance, but it’s actually a must-be, you will obviously miss a huge opportunity to improve the initial impressions of your clients. And in the opposite case, if you spend too much effort on a feature that ends up changing nothing in terms of performance and reception, that’s a huge waste that will eventually drag down your organization.

Leveraging Old Data Sets

This drives us to an important point if you don’t have preexisting data sets related to the specific product you’re developing, you may not be able to get a good overview of the different types of needs of its consumers. With that in mind, the initial development of the product’s very first version should be driven by a more experimental approach, although if you can draw on data from external sources, this can definitely be helpful as well.

And as you’re probably guessing, the data generated during this first run will be of critical importance later on if you want to continue working in that market and you want to make sure that your next product will be a noticeable improvement over the first iteration. How you’re going to record and store all that data is entirely dependent on the type of product you’re working on, and while some industries have established standards for data collection, retention and analysis, in some cases you’ll have to come up with your own custom implementation.

The Changing Nature of Needs

Another problem you may face in this context is that customers’ needs can change along with the design of the product itself. For example, you may unknowingly introduce a must-be feature, which then changes each customer’s perception of other aspects of your product. That way, in the long run, some features will move down in the table and will turn into indifferent needs, while something else might get promoted to a must-be feature.

That’s why it’s important to run regular market analysis on your customer and always be aware of their exact requirements and opinions. This is also necessary due to the volatile nature of most markets themselves, especially today in a time when the internet is making everything so dynamic and unpredictable.

The good news is that this system can also be very beneficial for collecting opinions and impressions from a large number of customers (both current and potential ones), and as long as you leverage your presence on the market correctly, you shouldn’t have any problems building a good model of your customers’ needs, and figuring out the exact direction to take your next product in.

Conclusion

The Kano model is a good framework for addressing many types of issues that arise in the development of a product, but it relies on preexisting data to make accurate decisions. If you’re developing an entirely new type of product or you don’t have much experience in its relevant market in the first place, you will need to do some groundwork before you can apply Kano effectively. Once you’ve gone over that barrier, you’ll find the model to work very well on a small scale as well as a large one, and it will give you a valuable insight into the way your company should work.

The post Can the Kano Model Be Used Effectively in Product Development? appeared first on 6sigma.

Statistical Process Control or SPC for short can help your organization progress in great strides when applied correctly, but it does come with a few caveats that you’ll need to observe. Most importantly, SPC requires you to spend a lot of time figuring out how exactly you’re going to collect your data, and proper data collection practices are the foundation of a good SPC implementation. Get that part right, and you won’t have to spend a lot of time figuring out the rest.

The Two Types of Data

There are two main types of data you’ll be dealing with in SPC attribute data and variable data. The main difference between them is that attribute data is based on attributes discrete, mutually exclusive properties. A product is good or bad, a condition is true or false, a certain percentage of products fall in some category, etc. The point is that attribute data can easily be grouped according to these values, allowing you to aggregate the data quite effectively.

On the other hand, variable data is continuous and not discrete. For example, it can measure some range between 0 and 1, the value of an analog control signal, and more. Variable data also tends to be more flexible in its application to your research, as it can give you a more objective overview of the current situation.

That’s because attribute data can be very susceptible to your own perception, and in the end it can be quite subjective. How do you define whether something is good or bad? Unless you have a very strict table of product requirements that covers every single aspect of the output (which is not a bad idea at some point anyway), this can be quite open to your own interpretation.

Combining Your Data Sets

Sometimes it can make sense to draw conclusions from more than one data set, including ones of different types. You can, for example, look for overlaps between an attribute data set that covers the final quality verdict for a product, and a variable data set that shows what signals a certain machine in the production chain has been receiving when each of those products arrived at the end of the line.

This can lead to the discovery of some interesting relationships between the work of some parts of the production line and the final output, in some cases giving you unpredictable results that you would have never guessed on your own. That’s one of the strong points of SPC, and the main reason why you should spend as much time as you can developing a good, reliable system for data collection and aggregation.

Aligning Old Data with New Discoveries

Sometimes quite often in fact it can be useful to compare some old findings with new data to see how some of your company’s processes have evolved over time. However, it’s not rare that you will have modified your data collection practices in some way from the last time that data was collected, resulting in incompatibility between the sets.

This can be avoided by setting up your data structures and databases to be flexible and easily adaptable from the beginning. Use a common standard across the board and try to implement everything in a modular way that leaves it open for change in the future. It doesn’t have to be hard to align data sets taken under different conditions, as long as this was planned from the beginning, and if you do it right from the start, you can easily keep your entire historic data useful for as long as the company itself exists. Which, as you’ll eventually find out, can be a huge benefit.

Conclusion

Proper data collection practices are a critical aspect of SPC, and something you’ll want to master as early as possible. Don’t rush the job thinking that you’ll just come back to fix things later the longer you wait between separate iterations, the more problematic it will become to compare different sets of data in the future, defeating the whole point of running a continuous statistical analysis on your organization’s work in the first place. On the other hand, when you do this right, it can save you tremendous amounts of effort in the long run.

The post Data Collecting in SPC: A Short Overview appeared first on 6sigma.

How Do We Use Machine Data Capture?

The types of machine data your computer systems collect will vary. It can range from the volume of satisfactory products to sell vs. rejected sub-standard ones to machine capacity effectiveness and utilization. Additionally, your systems should also monitor factors like machine cycles, production time, availability and reliability. Machine status is an equally critical component of MDC and denotes a whole spectrum of important production factors. These factors include primary and secondary time, breakdowns and maintenance, as well as service requirements.

But what happens to all this data? How do businesses use it? It’s simple. All your captured data travels through interfaces where your manufacturing execution system (MES) collects and collates it. Think of your MDC as the gold miner and the MES as the sieve through which they filter all the information they collect. The combined MDC data and MES findings allow you to use data effectively. Furthermore, your goal is to use your data to plan and push production in the direction you want it to go. That is, toward achieving optimum conditions. This allows for increased productivity and better results.

How Can Six Sigma Improve Data Collection?

You’re probably wondering, how does Six Sigma fit into the equation? Six Sigma improvement projects are an extremely reliable method for process improvement. It allows you to make the changes necessary for a strong data collection plan. Moreover, Six Sigma project leaders, usually Black Belts, employ DMAIC to define, measure, analyze, improve, and control data collection processes. There are several prerequisites you must meet for an effective data collection plan.

First, we have the pre-data collection steps. Ensure your project team defines your data collection goals. What data do you need? Equally, for what purpose will you use it? What insight will it offer and what do you wish to achieve through collecting it? Using tools like brainstorming, affinity diagrams, and root cause analysis can help generate answers to these questions.

The project team should then be able to agree on your plan’s methodology and operational definitions. Teamwork is always most effective here, and we recommend examining previous data to compare to the current. It’s essential that you determine whether past, present, and future data will factor into the data collection plan, plus what methodologies you are likely to use. Skipping this step will certainly deliver insufficient, if not deceptive, results. Finally, it is essential you ensure the repeatability, accuracy, stability, and reproducibility of your data collection and measurement. Once you have defined and planned your data collection process using the above data, you can then move forward. Black Belts can oversee implementation and to nudge it in the right direction, should you run into any obstacles.

The post Six Sigma and Business Analytics: Machine Data Capture appeared first on 6sigma.