How do you measure process improvement with data collection?

Process improvement clearly is not possible without data. It requires data to describe the current state of the process, the desired state of the process and the metrics needed to measure the improvement activities as well. However, the data used through out the phases of the process improvement must be complete, reliable and relevant to the process in question, otherwise the entire activity becomes an exercise in futility.

Foremost, you have to understand the current state regarding your Key Performance Indicators (KPI's). Are you leading or lagging the industry benchmarks, or somewhere in between? Then ask "What is the sense of urgency for improvement in my organization"? This second question is critically important to goal setting. If your organization lacks the resources, vision or fortitude to make improvements at the pace required, this has to be factored into the goal setting process. Misunderstanding the cultural will to improve has doomed many a well-intentioned continuous improvement endeavor to failure.

Although it’s tempting to gather a large quantity or large variety of data, it’s important to avoid making the data collection cumbersome or difficult. If it’s difficult to measure or takes a lot of work, it may not get done or if it is, be done poorly. Try to keep data collection as simple as possible.

Be sure to validate your testing equipment and methods before beginning data collection. A simple GRR or even less sophisticated methods can shed valuable insight before spending time, energy, and effort collecting noisy data.

Before collecting your data always ask yourself the question "What do I want to learn about the product or process?". Your answer to that question will help you take the best data possible.

Michael’s statement re-testing equipment validation is the most profound, important statement and the most overlooked aspect of data analysis. Your data are suspect until proven otherwise.

Attain with your end users input. Analyse for them with clear explanation on the interactive dashboards designed for them and then this provide the fuel to action. With each cycle the collaboration grows and you are on the journey to longterm improvements that being sustained benefit .

Define your measures before your assessment. Then you can determine if you have benchmark numbers to compare against. There's nothing worse than conducting an assessment to find out you have no way of measuring what happened in the past.

As you analyze your data always do a sanity check on what you are learning. Remember you are the master of the tools. They are not your master. Sometimes an unusual data point can throw off your results. Sometimes a hidden lurking variable might be the actual driving force. If the analysis is not making sense try to figure out why.

Often after you get some measure of data and start into analysis, you immediately start to gain insight into what is going on. This may make you realize you need different data. Do not be afraid to go back and change your measurement plan and adjust as you learn. Furthermore, once you have a clear root cause and an area of focus, resist the temptation to continue to gather more data just to check some boxes. Data collection and analysis is a lot of work, so follow the 80/20 rule when it comes to quantity and quality.

Always keep your clinical eye quite open!, and your professional criteria about deviations, if they are representatives or just under control parameters. Beginners somethimes use percentual tolerances criteria, when sometimes, these 5%-10%-15% of tolerance in deviations can give you the main response to your waste (muda). So, keep your eyes always fresh and open 👀, always ready to quesiton the results and deviations! 🦾😃

For me this is one of, if not the most important sections. The analyze segment here is key and will help guide your focus when looking at improving and standardising the process for the improve step. It's important to work on what you should be working on and not what you want to. As long as the data is there, you can't be biased! Lastly, make the data as visual and easy to interpret as possible. This makes buy in from your stakeholders so much easier.

At the improve phase it’s very important to get stakeholder buy in. This includes shop floor people. Having their understanding and support will make the monitoring phase much easier, and ensures open and honest feedback.

While improving the product or process always consider the control aspect as well. The best way to make sure things stay fixed is to implement controls at the same time as you implement the improvements. Waiting until the end of a project to think about how to keep the product or process fixed results in poor control plans.

Having those most involved with the process is critical for buy-in, as well as for ownership when transferring to CONTROL phase.

After evaluating each potential solution based on factors like feasibility, potential impact, cost, and alignment with the goals, always consider potential risks and develop mitigation strategies into a detailed plan for implementation. Implement the chosen solution on a small scale and in a controlled environment to test its effectiveness. Make necessary iterations and modifications as needed to enhance the solution's effectiveness. Ensure effective communication and provide adequate training and support to relevant personnel. Implement standard operating procedures, quality control mechanisms, and performance monitoring systems to ensure the continued success of the solution, and audits are recommended during the first few months.

I’ve been looking into this article about how data collection is important for process improvement, and I think point #4 covers it best. Once you have the data, you need to define targets and monitor the process at regular intervals. You can also break the process into subgroups to more accurately set targets. Regular feedback is essential to sustaining the process of improvement.

Make sure the final report has an element of celebration. Make sure those who helped you feel appreciated. This helps drive the culture of Continuous Improvement and helps make sure when you need help again that people will be happy to help.

Communication throughout a project at all levels is really important. Use different methods to communicate results, encourage feedback at every stage and respond to questions. The finished results are continued by a wider team, recognise that.

Make sure you update your process documentation. Especially the tools that will help you find the next opportunity. Things like the Value Stream Map and Process Map. Remember the best Value Stream and Process maps are living documents that are always up to date to show us our opportunities to Continuously Improve.

Apples to Apples: Make sure you are measuring the same thing after improvement and consider any secondary effects (positive and negative). Ensure your future state improves on the exact problem you defined using the exact same metrics. Be honest and understand what is working and what isn't. Highlight exceptional days and also the disasters (dig deep to find why they occured). The name of the game is being consistently better than yesterday without hurting yourself in another area. It's not so much a trade off as it is an upgrade. A careful review at the end can yield insights that will lead you to the next project!

All too often, we measure and score card performance by only the mean of the process when often process outputs like cycle time, production output, and the like are non-normal in distribution. In those cases what drives the mean of the process is not always the same as what drives the process variation. We forget that the true evil is variation. Too often we measure success solely by measuring improvements in the mean, and we forget that when you focus on reducing variation first you get not only a more predictable process performance you also get a simultaneous shift in the process mean as well. Focus on variation first!

Variation is the evil of improvements. At Toyota they taught us Kaizen and then sustainment. Sustainment is about variation in Cycle Time. You need to improve and sustain. That is the biggest challenge in operations. It’s not about Lean, OPeX, or 6Sigma. It’s about a culture and sustaining change. Most zealots as I call them miss this culture piece. It is not soft things as most people think, it’s about how you operate and make things happen. Honestly most people think they know the Toyota Production System but until you truly live it you really don’t know the core. I’ve had so many people say I spent time with the trainers but they focus on tools. It’s not about tools. Get that into your head. If you understand that you will win.