Modern industrial and manufacturing processes bring with them tremendous complexity. Today, most factories rely on elaborate supply chains, divisions of labor, and technologies to produce their goods. Amid this sea of inputs, people, and technology, it can be challenging for teams to find operating efficiency gaps, let alone improvements.
Enter the Management Execution System (MES). Sitting atop existing capabilities such as Supervisory Control and Data Acquisition (SCADA) solutions, an MES can deliver a complete overview of an industrial process’s lifecycle. By providing data-driven insights across every step of a production process — whether procurement, logistics, labor, maintenance, quality control, or beyond — the benefits of MES can be tremendous in boosting your operating efficiency.
In previous installments of this in-depth series, we introduced what an MES is and why it’s so effective at optimizing processes. Here, we’ll explore how you can ensure that your MES provides valuable insights to your team.
Generating Outputs Beyond Raw Data
Our last article covered the distinction between the two types of information generated by an MES: overall equipment effectiveness (OEE) metrics and statistical process controls (SPCs).
To summarize, we can say OEE data describes the relative efficiency of an industrial process. An SPC, by contrast, prescribes improvements for an operation. If you’ve implemented your MES correctly, you can obtain OEE metrics and SPCs in a user-friendly format to share with other team members.
By contrast, a “bad” MES implementation doesn’t present its output in an easily accessible or actionable way. Instead, it just compiles raw data and, at best, offers some basic inferences regarding how certain variables in an industrial process relate to one another. So how exactly do you steer clear of “bad” MES implementation?
Avoiding Over-Reliance on Historical Data
If you gather weeks, months, or quarters of historical data and feed it to your MES, there’s a good chance it will fall short. Let’s look at an example to see why.
Imagine your warehouse workers all catch a common cold. Many employees may not notice they’re ill, and even those who do will try to work through it. But despite that, there’s a strong chance of a noticeable productivity drop among the warehouse team. The secondary effect of this could be an organization-wide reduction in OEE.
Most organizations don’t systematically log the spread of illnesses that aren’t severe enough for absence. If an MES were to analyze data from this period well after the fact, it’s likely to come to invalid conclusions about why the OEE fell. An MES may infer, for example, that the reduced output was due to recent changes in the warehouse’s layout.
This example shows that there are events an MES can’t perceive. Many more situations have contextual information attached, which changes what we should reasonably infer. Much of this will be missed when submitting historical data in bulk to an MES.
Instead, you should deploy your MES to oversee a live production environment. What does this look like?
Practicing MES Hygiene
To start, you must connect an MES in a live production environment to every endpoint in the manufacturing lifecycle. One example of such an endpoint is the programmable logic controllers (PLCs) that control actuators on production lines. However, given the full-view goals of an MES, these endpoints go beyond just the assembly line proper. An MES must also draw from live supply chain information, maintenance records, labor data, and beyond.
An MES can assess pain points causing shortfalls in OEE on a daily basis. Your team can then examine this input as it comes in, and manually intervene when the MES’s decisions are at odds with facts or contextual information.
This initial dataset on errors or OEE bottlenecks is your starting point in delivering actionable insights. You can funnel this information on errors into reporting, where it’s joined by a second element: cause codes. As the name suggests, cause codes are tags that identify a particular cause of an error or bottleneck. They offer straightforward means to categorize issues and rank the severity of their impact on an industrial process.
Cause codes are customized for each industry, use case, and setting to reflect the facts on the ground. But they don’t just offer an easily identifiable and organizable way to sort and categorize errors. Cause codes also allow for user-friendly outputs from your organization’s MES and a means to triage which improvements should come first.
Working With ICA to Best Leverage MES Data
Leveraging an MES is not a matter of “plug and play” — teams have to strike the right balance to capitalize on the capability to gather data and report insights automatically, while also ensuring it generates readable and actionable outputs.
The need to strike this balance is why you should manually review an MES’s day-to-day inferences. You must also provide the means to productively leverage MES insights by choosing the proper cause codes for your reporting framework. From this, you can generate the usable and valuable reports your organization needs.
But what are the sorts of shortcomings that MES reports are best-suited to remedy? Stay tuned for the next article in this series, where we cover the common operational pitfalls an MES can help you avoid.
ICA Engineering has unparalleled expertise as a partner in maximizing the benefits of MES while also helping you strike a balance in generating actionable insights. Contact us today to learn more about how we can help you improve your operating efficiency through a management execution system.