In manufacturing, what you don’t know is likely hurting you – your efficiency, your productivity, your culture, your bottom line.
But you don’t know what you don’t know!
That’s where data analytics comes in. It brings light to what you don’t know so you can take action to fix it. On a manufacturing line, the challenges could be process bottlenecks, operator performance, product difficulties, scheduling, over-cycle occurrences, or other hidden inefficiencies.
Simply put, data analytics is the application of statistics on data in order to analyze and improve practices and complex manufacturing processes. And, given the right software and professional guidance, data analytics can become a natural extension of how you think about your plant.
Action rooted in knowledge is the path to manufacturing excellence. Let’s break down this philosophy into simple steps:
As opposed to “big data,” which attempts to establish relationships within a mass of dissimilar data, data analytics is a granular approach to diagnosing and correcting flaws in a process with a specific goal in mind from the beginning. This narrow focus is really the key to ensuring the effort results in significant time and money savings.
When data is aggregated even over a short amount of time (say, 30 days) and is then analyzed, the data analytics software uncovers not only numerous insights but also unexpected surprises and opportunities to increase yield.
Capturing the right data is always priority one. To address specific problems, you need to obtain certain required data. First, determine what types of analytics and associated processes will make an immediate impact, and then define the data requirements and a roadmap for future needs.
Two data sets can be quickly created: product (everything that happened regarding a specific product) and manufacturing (stations and operators). So, both learning about your products and how you make them ultimately helps increase production.
A big challenge for data science professionals is ensuring that insights are actionable. Too often, little thought is given to how users will digest and use the analysis. Once that’s addressed, determining action steps for various positions becomes easier.
It’s important to deliver analyses to each audience in a way that’s clear and not cluttered with information that they find irrelevant or confusing. From there, the insights can inspire different actions with different groups to help fulfill their responsibilities:
Operators use data analytics to…
Line managers use data analytics to…
Plant managers use data analytics to…
Modern manufacturers are using manufacturing execution systems (MES) to uncover data analytics. MES solutions identify unique problems and help plan action steps, whether those relate to machines, operators, or both.
The following example shares the processes of leading MES solution PINpoint V5. The chosen manufacturing problem is “operator performance variability,” one of four challenges identified by plant management.
The company’s main line was producing 430 units a day but had a goal of 525, and the presumed cause was operator turnover, with new operators not being able to meet cycle times.
Now, let’s dig in...
Investigation included inspecting data, conducting analytics over time, analyzing the operator/product relationship, and visiting each station. An MES analytics report showed average cycle times by station and by operator, and a dashboard showed performance differences between operators.
Let’s dig deeper...
Uncovered knowledge included an overall look at productivity as well as issues at each station and with each operator. It was clear that a specific product on a certain station was challenging for a majority of operators.
Let’s keep digging...
Action steps were determined by management based on what the analytics uncovered. A “best practice” training approach was established with better operators training others, one operator was removed from the line, and a tool was redesigned to accommodate left- and right-handed operators.
Results included line output improving by 12%, production losses reducing by 55%, and units produced increasing by an average of 51 per shift, which is equal to an extra 2.4 shifts per month!
That’s just one high-level example of an MES’s capabilities. Other challenges tackled by the software for this customer included bottleneck stations, hidden inefficiencies, and over-cycle occurrences. And each of those included details such as photos, charts/graphs, dashboards, and many more tools.
With the right approach, it’s easy to get everyone on board with using data analytics.
Remember, people don’t know what they don’t know. When “action rooted in knowledge” is presented as the path to manufacturing excellence, it’s an eye-opener from top to bottom.
Typically, an orientation workshop for experts and management conveys a basic understanding of analytics’ possibilities and identifies some use cases. An analytics team can then start working with line managers to determine analytics goals.
Keep in mind the three main types of employees that each require a slightly different approach: skeptical (not sure this “data analytics” thing will help at all), open-minded (willing to listen and hear evidence), and believer (already on board; understands a new approach could be helpful).
Manufacturing is a high-stakes game. There are many operators, many stations, and many costly hours involved. Understanding it all can only come from analytics, which captures and analyzes the data to provide insights. Without MES software, you’re just guessing.
Instead of being picked apart by one small inefficiency after another, you’re making real changes quickly (within 30 days). You’ll uncover what you don’t know and get an accurate description of trends and productivity.
Even just raising productivity by a few percentage points will pay for your investment in an MES solution.
Want to dig deeper? Contact PINpoint Information Systems. We can talk about data analytics for days, but we promise to stay focused on your manufacturing line needs. Complete this form or call 905-639-8787.