top of page
  • eLogic

Scrap What You Thought You Knew About Analytics

The Scenario

Your operators and supervisors know there’s a problem with product quality. At best, you’re not seeing it until after the fact in a monthly trend report. Or worse, not until someone yells that there’s a proverbial fire. Even more challenging is when, despite all their best intentions, your people try to adjust on the fly without knowing the bigger picture of how their work is impacting the entire organization. So how do you go about having the visibility needed to improve quality, decrease scrap, and thereby reduce costs? By capturing the information, you need to generate right-time analytics that give actionable insights.

So, how do you begin?

Define the Problem

The first step is to clearly define the problem(s) you are trying to solve. That sounds easy enough but there are things to take into consideration. This is not to say that you should spend months and months trying to get the perfect requirements. In fact, if you take that approach, the original problem you were looking to solve may have morphed into something entirely different. What should you do instead? Work with the key stakeholders to understand what they are seeing. Perhaps you’re seeing a high amount of scrap in the manufacturing process. Test that understanding and confirm, in fact, there is a high percentage of scrap quantity (or dollars) as a result of a grinding process. Then move from the theoretical into the practical and walk the floor to see the issue in the native environment, live and in living color.

Gather Requirements

Great, now that you have confirmed and know what the problem is… what’s next? Work with those same people to understand what information they want to view and how they plan to assess whether a problem is really an issue or, if it is, that the issue is being addressed. Some might call these requirements, use cases, or user stories. Whatever you call it, the goal is to have short, simple descriptions of the information they are looking for and to associate the functionality to the appropriate audience’s dashboard. When you investigate, you realize that to account for the higher amount of scrap, the department is producing a high quantity of product in the hopes to produce the necessary order quantity that meet quality standards. A certain amount of scrap is to be expected. But is this an isolated issue, or a recurring problem?

To figure that out, it’s best to go about the business of creating the secret sauce of defining the metrics and their components. Reports have their time and place, but they’re not the same as metrics that drive analytics and give you the edge over your competition. So to that end, if you are trying to get visibility into being able to improve quality, decrease scrap, and reduce costs, think about the scenarios, data fields, and business rules that will highlight an issue. Sample metrics may include daily and monthly scrap quantities and dollars, with the ability to segment that data by department in order to pinpoint where the focus should be on the production life cycle. What’s more is making sure you determine the timeframe and proactive alerting methods, such as emails and texts, so that you can take the appropriate preventative action or, at the very least, can react quickly when the problem occurs. Through capturing the information you need to generate right-time analytics, you will be able to keep and, better yet, expand your customer base.

Determine the Dashboard Metrics

The best practice for displaying this data is to provide views at the enterprise, plant, and departmental levels. This allows executives, plant managers, department heads, and supervisors to easily aggregate and drill down to see which plants or departments are contributing to this scrap issue. This is where your journey begins to get a thorough root cause analysis. But the approach to really bust through the finish line and win the race is key. Take a similar approach to requirements by not spending a bunch of time doing work and presenting the end result for review. Build the metrics, graphic elements, and dashboards in an iterative fashion with very frequent review intervals. Think Agile in that regard. More often than not, the audience won’t know what will work best for them until they see the live metrics in action.


If you approach your production issues this way, you will be able to proactively avoid and address them while avoiding frustration on the parts of you and your customer. Plus, an added benefit will be spurring leadership onto new thinking, with improved customer service and analytics that are meaningful and actionable. This is truly how you make quality king!

bottom of page