This blog is the second in a two-part series. The first article, “Achieving Magical Gains in Manufacturing Efficiency by Measuring Overall Equipment Effectiveness (OEE) - Part 1 of 2” explained how to achieve manufacturing efficiency using OEE and an additional metric called loss deployment. We'll take the conversation further by discussing how to achieve predictive awareness in manufacturing efficiency.

When manufacturing companies approach ICONICS and our system integrator partners, besides looking for improvements in their OEE, they want some indication that their performance, their efficiency, may go wrong prior to it happening. They want some sort of predictive awareness. An additional part of a manufacturing efficiency project is not just to find insights into current operational challenges to improve efficiency but to go beyond this and reach toward a predictive awareness of that efficiency. For a bit of background, I’ll explain OEE first. 

A Quick Down-to-Earth Refresher on Overall Equipment Effectiveness (OEE)  

In most manufacturing efficiency efforts, as explained in part 1 of this series, we start with Overall Equipment Effectiveness (OEE) as the first measurement of data and as a starting point to focus our efforts on improving efficiency. Just to refresh your memory, OEE is made up of the monitoring of a machine’s availability (which is uptime versus downtime), multiplied by its performance (which is the speed of a process or a machine compared to the designed speed of that machine), multiplied by the quality of the product produced by that machine (which is good parts versus bad parts – the sellable parts versus the scrap). The uptime of the machine, the performance of the machine, and the quality of the machine multiplied together tells us the overall efficiency of that machine. 

Quite often in some industries, a fourth pillar is added to OEE, which takes into account the material used. This is the material that was delivered such as packaged or palletized goods compared to the raw material that entered the process. This fourth pillar will often come into play, for example, in the food and beverage industry or anything to do with plastics or packaging in manufacturing, where in addition to knowing if a line is available, we also want to know the performance of the line. We want to know the quality of the product in addition to knowing the efficiency with which we utilized the material that went into making it. If we take this further, we can weigh our OEE score against cost by looking at the cost of operating a line in terms of the capital equipment cost, the material cost, the labor cost, and consumable costs. It's important to look at not only the efficiency of a line but also the cost of labor to achieve that output. 

After considering these metrics, the next level in a manufacturing efficiency project would be to look for ways to achieve predictive analysis on improving our performance, our use of raw materials, the uptime of the machine, the quality of the product, and so on. To understand how to get to this point, it’s important to have some knowledge of the concept of DIKW or the process of turning data into information, then knowledge, and then wisdom. 

Following the Data, Information, Knowledge, Wisdom (DIKW) Path 

The concept called DIKW (Data, Information, Knowledge, Wisdom) allows us to think about our data collection or data analysis project in terms of starting with the cost effective and organized collection of a system’s raw data and then turning that data into information by applying correlation to it. For example, if we take data and associate it with a given line or a given machine, that data becomes information which is associated with an asset. If I associate that data with a given shift, operator, batch of product, or type of product, I've now taken that efficiently organized raw data and turned it into information by associating it with the product or the machine from which it came. The next step is to think about reaching toward knowledge, which is taking the information and plotting it so that we can learn from it. For example, let’s say we have two products: the red and the blue product. We may look at a chart that shows us a comparison of our efficiency when running the blue product versus when running the red product. And we may gain some knowledge by looking at that information. For instance, we may see that the process of running one product is more efficient than the process of running the other product. This process points of data, information, and knowledge are where most plants are today. But we want to go further to reach toward wisdom, which is a more predictive understanding of what's coming next, and that is based on the data in front of us. 

To achieve wisdom, you need two keys to help get from knowledge to wisdom. First, you need insight. Some folks might be inclined to jump straight into artificial intelligence (AI) and machine learning (ML) as tools for obtaining insight to get from knowledge to wisdom, but before diving too deeply into AI and ML, our recommendation is to take a step back and consider the experience and knowledge that is already well within reach to get to predictive awareness: your employees’ experience and knowledge. Their expertise and know-how will help you predict maintenance requirements and performance and quality losses before you take on the task of applying artificial intelligence and machine learning on top of that. The other key is to be able to accurately measure (and track) the impact of your investment. If you make an investment in time and technology to jump from your current knowledge-based system into a wisdom-based system, it is essential that you can show a return on that investment. And the way to measure that impact is to go back and look at the change in OEE of your efficiency. Did your move from knowledge to wisdom, in fact, have a positive impact on OEE and therefore justify the investment that was made? These considerations and this approach can lead to predictive awareness of efficiency: predictive maintenance, predictive performance, and predictive quality. I imagine at this point you want to see how by looking at some examples.  

Well, you can by watching “Achieving Magical Gains in Manufacturing Efficiency" (a session from our Transform 360 " webcast series). In this webcast, President of Data Acuity Inc. Jim Desrosiers, an expert on OEE and manufacturing efficiency projects, clearly explains how to establish predictive awareness in manufacturing efficiency and gives examples for attaining predictive maintenance, predictive performance, and predictive quality. If you haven’t already seen it, feel free to go watch that now and then check back with me. I’ll wait. You back? Okay, great! Now that you know you can achieve gains in manufacturing efficiency along with predictive awareness, you might be wondering how to start such a project. Here are some tips: 

What to Consider When Embarking on a Manufacturing Efficiency Project 

1. Start at the top (data consumption), not at the bottom (data source). 

Engineers tend to focus on one problem at a time, one technical detail at a time. However, if you’re starting a manufacturing efficiency project, it's important to put aside that natural tendency to focus on one technical detail at a time, and instead to go to the very top level, to the end result of the project to consider the data consumption, or how that data should be viewed, not the data source(s) available to you. So, you're going to be thinking about the screens, the reports, and the web pages you're trying to pull together that will give insight into your efficiency and lead toward predictive awareness. And then you can work down toward the data source(s) that will get you there. 

2. Organize your effort more around subject area and job roles rather than a natural tendency to organize it around your existing tools.

If you have a labor shortage within the maintenance department, it’s important to focus your efforts around those subject areas and those job roles, as opposed to what is the current tool set in front of you. At the end of the day, your people are your most valuable assets, so it is important to make sure your efforts incorporate their expertise and knowledge while earning their trust and buy-in with respect to the tools you are leveraging to help them do their jobs more efficiently.

3. Understand the long-term possibilities but move forward with a realistic and highly confident first step.

Typically, you want to consider the types of data consumption you'd like to achieve to gain more efficiency within your plant over the initial project period which is commonly three to five years. And it's extremely valuable to focus on a single small project to get started. Therefore, consider where you want to get to so that that knowledge is in the back of your head as you're designing a system. But to take that first step, you need to define a small project with a high chance of success to get started on.

4. Think about the data structures that consider future correlations.

If you’ve thought about the data consumption you need for your project, the graphics and reports that you want to implement over the course of a three to five year plan as you're building your smaller first step, you may want to start building the data structures, data tables, to house the appropriate quality and performance information that are going to help you with those data correlations.

5. Consider some level of Agile Project Management.

It doesn't have to be a formal agile structure, but you just want to keep in mind that when you're thinking about an initial scope of work or an initial design of a project, in all likelihood, the direction of the project can and almost undoubtedly will change. As you gather data and as you analyze data, you're going to learn what the real losses of efficiency are and then perhaps you will need to take a left or a right and head down a different path. Hence, it’s essential to be ready and prepared to adapt quickly by having an agile project structure.

​6. Perform a formalized assessment if there is time. 

Doing a formalized assessment prior to building a manufacturing efficiency data system can be a tremendous time savings. In the real world, we understand it can be hard to put aside time and budget for a formalized assessment. We almost always need to take on that first project just to get moving. But if you have the time, then use it for an assessment.

7. Take a phased approach to deployment.

You need to be prepared for the fact that manufacturing efficiency projects are usually not a “one off” project. These projects will take on a life of their own with iterative processes of evolution and improvement. Thus, it’s very reasonable (and we think smart!) to do a phased approach to a project, to complete phase one to show some results and then move on to phase two to show more results. In this way, a manufacturing efficiency project can grow and move forward and provide returns indefinitely.

8. Keep an eye on ROI.

It’s extremely important to monitor the impact of your work to make sure that your return exceeded the cost or the investment of a data collection project. So, keep an eye on the return on investment.

Talk to the Experts When You’re Ready to Start Your Manufacturing Efficiency Project

This two-part blog series was aimed at giving you an overview of the gains you can expect in a manufacturing efficiency project, and what to consider before starting such a project. However, you don’t have to go it alone. There are experts with decades of experience and an immense amount of expertise to help and guide you, companies like ICONICS and Data Acuity. You can fully grasp the value these companies can bring to you and your company by watching “Achieving Magical Gains in Manufacturing Efficiency”. In this webcast, ICONICS Solution Sales Engineering Supervisor Jotham Kildea explains the key elements of a successful manufacturing efficiency project and the technology and experience ICONICS can bring to such a project while President of Data Acuity Jim Desrosiers explains the how’s and why’s of a manufacturing efficiency project, including a great manufacturing efficiency use case from our customer Catania Oils and other valuable examples.

Given all this, there really isn’t any reason not to get on with your manufacturing efficiency project to improve your operational efficiency. Good business practice is based on continuously improving manufacturing efficiency, and the time is now to start working towards achieving those gains.