Ep. 18 Why Configurability in MES Is the Key to Scalable, Innovative Manufacturing

Click below to listen on the LinkedIn Platform

Remember to follow our page for more updates!

[A: 00:08.0]
Picture this. It's quarter, past six in the morning, you're the production supervisor, just hitting the shop floor. Straight away, there's a rush order landing. One of your key machines is acting up and oh, a supplier just called Last minute material change. So, yesterday's perfect plan Pretty much useless now.

[A: 00:24.2]
What happens next really hinges on whether your systems can pivot, like right now. Our guide for this deep dive is an incredibly insightful article. It's titled "Why Configurability in MES is the Key to Scalable, Innovative Manufacturing" Now, this piece comes from MASS Group and it's written by their CTO, Paul Vasallo.

[A: 00:44.0]
He's known for, well.. guiding complex technical initiatives and delivering scalable solutions for some big names in manufacturing. So, the big question we're tackling is this why is configurability not just some nice to have feature anymore? Why is it absolutely critical for manufacturers if they want to thrive in today's let's face it fast paced, constantly changing world?

[A: 01:03.4]
Our goal: to figure out how this idea of a configurable MES actually helps you deal with that kind of daily, well, chaos. Right. And it's important to understand

[B: 01:12.2]
configurability isn't just some marketing buzzword. It's really an architectural decision. It means you treat things like data models, workflows, even the screens people use as actual managed assets. They have versions You can test them, you can push them live without rewriting the core software itself And that connects to a basic truth on the floor.

[B: 01:29.3]
Processes just change way faster than code should. Configurability bridges that gap

[A: 01:33.8]
That adaptability sounds ideal.

[B: 01:36.1]
Think of it like this A configurable MES treats the fundamental bits, data models, workflows, user interfaces as first class versioned assets

[A: 01:47.4]
Okay, versioned assets What does that mean practically?

[B: 01:49.7]
It means you can compose them, test them, promote changes to them all without rewriting the core code of the system itself. And in a market that's constantly dealing with complexity, tough regulations, endless product variations, well, that architectural difference is what genuinel y turns that daily disruption,

[A: 02:05.9]
this 6AM Chaos

[B: 02:07.5]
into actual momentum for the operation.

[A: 02:09.2]
Yeah

[B: 02:09.5]
it lets manufacturers be proactive, not just constantly reacting.

[A: 02:13.4]
That's a really powerful idea. And it reminds me of something you mentioned earlier, that line processes change faster than code should. I mean, for anyone in manufacturing, that must really resonate. Your reality on the floor is always outpacing your IT systems. So if that's the truth how did the industry even arrive here?

[A: 02:30.4]
What's the story behind MES evolution?

[B: 02:32.9]
Right, it's definitely been a journey. We can really trace this through, let's say three distinct eras. First, you have custom built systems. These were, incredibly detailed. They mirrored a specific plant's reality perfectly, often deeply integrated.

[B: 02:48.3]
But the downside, and it was a big one, was that every little change, a new product variant, even a small routing tweak, became this painful, slow code change.

[A: 02:58.2]
Okay. So very tailored, but very rigid.

[B: 03:00.4]
Exactly. It led to stretched timelines for any modification upgrades were incredibly difficult, often skipped, and you couldn't easily take what worked in one plant and use it in another. The spoke. Yes, but brittle.

[A: 03:11.5]
So after those custom systems, what came next? Where did manufacturers turn?

[B: 03:15.0]
Well, that led naturally into Era 2 COT suites That's commercial-off-the-shelf software. These definitely sped things up. Initially they came with strong baseline functionalities quicker deployment paths and improvement for sure

[A: 03:28.1]
there's a but coming I sense

[B: 03:29.6]
there is Flexibility was often tied pretty tightly to the vendor's release schedule. And here's the real kicker. If you wanted to extend these COTS systems to perfectly match your unique process.

[A: 03:41.3]
You needed custom code again.

[B: 03:42.8]
You often ended up right back there, reintroducing custom code and all the technical debt that Era one was known for. You bought a product, but making it truly yours often put you back in that customization trap.

[A: 03:54.5]
Okay, so that brings us to now: Era 3

[B: 03:57.4]
Era 3: configurable or you could say composable platforms This is the current modern, pattern What defines this era is low code tooling built on top of a stable, reliable core system. With these systems, the MES platform itself becomes, as the article puts it an engine of change, not a barrier to it

[A: 04:16.2]
An engine of change I like that.

[B: 04:17.7]
Yeah, teams can actually tailor the workflows, the data definitions, the user interfaces to match how their operations evolve over time. And crucially, they can do all this while staying upgradable, staying agile. It's about building a system that grows with the business instead of holding it back.

[A: 04:34.4]
So connecting this back, that core idea processes changing faster than code That's the real driver for this shift to configurability.

[B: 04:43.9]
Absolutely. That simple operational truth is the fundamental reason. It explains why configurability isn't just nice, it's vital. And when you embrace this configurable approach, the benefits are pretty profound.

[A: 04:56.2]
Okay, let's talk benefits beyond just saying agility. What are the tangible impacts? What does a manufacturer actually gain?

[B: 05:02.6]
Right. Well, first there's faster time to value. This is huge. Instead of those massive 12 to 18 month big bang rollouts where you try to do everything at once

[A: 05:10.5]
Which sounds daunting.

[B: 05:11.7]
It is. Instead, organizations can start small, maybe just focus on electronic work instructions and basic traceability first. Then maybe add SPC, statistical process control for quality or scheduling or electronic records later. Value starts hitting the bottom line in weeks, maybe a couple of months, not quarters or years.

[B: 05:31.9]
Imagine seeing real improvements that quickly.

[A: 05:34.6]
That feels almost revolutionary compared to the old ways. An iterative, more manageable approach. What about the cost over the long haul?

[B: 05:42.3]
Good question. That leads to the second benefit. Reduced total cost of ownership. When your specific business logic, your workflows, your rules, lies as configuration on top of the core software, well, upgrades stay as just upgrades.

[A: 05:56.8]
Ah, you're not rewriting custom stuff every time the vendor issues a patch.

[B: 05:59.7]
Exactly. You avoid those brittle customizations, those weird one off UI changes tied to a single

[A: 06:05.2]
developer, and those really expensive rewrites that used to plague systems. Every time a process changed, your spending actually aligns with your operational needs and goals, avoiding those nasty hidden costs later.

[B: 06:15.7]
And for larger companies, maybe with multiple factories, how does configurability help with standardization? That's where enhanced scale and governance comes in. Configurability lets you use a template LED approach, define your global standards, master data models, core golden workflows just once.

[B: 06:34.3]
Then you can easily localize the specifics for each site. Things like equipment connections, local calendars, specific compliance rules. So you get this single instance visibility across the whole enterprise, but you still maintain that crucial local agility.

[A: 06:49.0]
Best of both worlds then, consistency and local responsiveness.

[B: 06:52.2]
Precisely.

[A: 06:53.0]
That template idea sounds like it could dramatically Speed up bringing in new tech too, like AI maybe

[B: 06:58.0]
You're right on track. That brings us to continuous innovation Because these configurable systems have open APIs, event streams, standard connectors. They stay in sync with your other critical enterprise systems, your ERP, PLM, LIMS, QMS systems historians, even devices right on the edge on the factory floor.

[A: 07:15.0]
So they talk to each other easily.

[B: 07:16.7]
Yes, and that seamless connection naturally invites advanced capabilities. Think AI assisted scheduling or anomaly detection. Flagging weird process behavior or closed loop quality initiatives. Feeding data back automatically. And you can bring these in without having to rip and replace your whole MES platform.

[B: 07:34.6]
It genuinely helps future proof your operations.

[A: 07:37.1]
Okay, that's compelling. But the actual process of implementing these changes. Does it still feel like a massive high risk project?

[B: 07:45.6]
It doesn't have to. The article also highlights phased low risk delivery. Because you can take this iterative disciplined path, it really stabilizes adoption and manages the change much better. Each implementation wave builds on and reuses what worked in the last one. That shrinks the effort needed for the next production line or the next plant.

[B: 08:03.8]
It's a build on success model.

[A: 08:05.4]
Makes sense. And you mentioned real time data earlier

[B: 08:08.3]
Right. Real time decisions, configurable dashboards and rules engines are key here. They can surface those critical signals Your work-in-process (WIP) overall equipment effectiveness (OEE) yield rates, SPC alerts right in context, right when they happen. This allows for immediate actions, triggering holds, rerouting production, escalating issues as events unfold.

[B: 08:29.8]
Not waiting until the end of a shift when the damage might be done.

[A: 08:32.8]
Closing that feedback loop faster.

[B: 08:34.4]
Exactly. And finally there's enterprise coherence. MES, when done right with configurability, truly closes that gap between the planning systems like ERP and the reality down on the shop floor. It keeps orders, bills of material, inventory levels all synchronized. And it maintains that critical cradle to ship product genealogy.

[B: 08:52.5]
This improves compliance, sure, but also just the overall operational flow. It's about having one accurate version of the truth driving everything.

[A: 08:59.5]
You know what's really striking is how these benefits seem to stack up. It's not just one thing, it's a whole collection of advantages. It really feels like it empowers manufacturers to take that daily disruption, the chaos, and actually turn it into a competitive edge.

[A: 09:14.9]
It does sound transformative, but I imagine for many manufacturers just the thought of a system overhaul is, well, scary. What's the biggest hurdle mentally for companies embracing this?

[B: 09:26.6]
Yeah, often it is just the sheer thought of change and maybe the fear of repeating past failures with systems that ended up being too rigid. That's why really understanding what configurable means, you know, down at the technical level is so important. It helps demystify it all and shows how it can operate more like a product, not a project.

[A: 09:43.4]
Okay, so let's get into that. What are these technical capabilities that make configurability work? What's under the hood?

[B: 09:49.2]
All right, first there's typically a data model composer. This is what lets you define your products, materials, equipment, all their attributes, the production routes, even the genealogy rules without needing database schema rewrites.

[A: 10:00.5]
Okay, so it's kind of like building your product database with Lego blocks instead of needing a master carpenter every time you want to add a shelf. That sounds way different from old rigid systems.

[B: 10:10.4]
That's a great analogy. Exactly. And that flexibility carries over into how things run with a, workflow and rules engine. This lets you orchestrate your production routes, holds how the system responds to SPC events, approvals, E signatures, escalations, all as low-code, versioned artifacts, things you can manage easily.

[A: 10:30.1]
Low code, so not deep programming required?

[B: 10:32.6]
Generally, yes. Often drag and drop or simple configuration to define that process logic.

[A: 10:37.6]
But what about the people actually using it on the floor, the operators?

[B: 10:41.4]
That's where a UI UX builder comes in. It lets you tailor the interfaces, the checklists, the forms, the dashboards for specific roles or workstations. Again without writing custom code. So the operator on line one sees exactly what they need to see. And maybe a supervisor on line five gets a completely different view tailored for them.

[B: 10:59.5]
And it didn't have to build two separate applications.

[A: 11:02.3]
That sounds huge for getting people to actually use the system. Yeah, what about connecting it to everything else? ERP, PLM?

[B: 11:08.6]
Yep, the integration layer is crucial. It uses standards based APIs and eventing technology, plus often pre built connectors for those other enterprise systems and also industrial equipment. And built to communicate. And very importantly, there are templates and promotion paths. This means your configurations, your workflows, your data models are managed just like software code, meaning they have audit trails, you can roll back changes, they move through controlled environments.

[B: 11:33.8]
Development, testing, production. This gives you control and ensures quality.

[A: 11:38.3]
And for regulated industries, pharma, aerospace

[B: 11:40.9]
Absolutely covered Built-in security and compliance features are standard things like role based access control, E-records, E-signatures, validated change control processes It's all designed to meet those tough requirements. And finally, you usually have flexible deployment options on premise, cloud, hybrid setups, often with single instance control, even across multiple sites, which helps manage those configurations centrally.

[B: 12:06.6]
Lots of architectural flexibility there.

[A: 12:08.6]
Okay, that technical breakdown really helps clarify things To make it even more real, the article includes a great case study. from paper to digital in 90 days It talks about a high mix engineer-to-order plant They were really struggling with old school paper travelers.

[A: 12:24.6]
Lots of manual tracking and dealing with too many defects caught late in the process. Their goal wasn't just let's do MES, it was specific. Reduce rework, tighten up their genealogy tracking and shorten those quote to ship times.

[B: 12:37.9]
And the results were pretty impressive, weren't they? The speed was key.

[A: 12:41.0]
Absolutely. It was rapid and phased. By week two they had a basic operator interface up and running, replacing paper on one pilot line just two weeks. By week five, their bills and material were cinched up with the shop floor. They had serialized WIP tracking, knowing where every unit was.

[A: 12:58.0]
And SPC alerts were live flagging deviations immediately. Fast forward to week eight. Dashboards were making bottlenecks clearly visible. This let supervisors actually tune routings and constraints right there on the floor in near real time. And by week 12, so just three months, supplier certificates were flowing in digitally and audit prep, which used to be this massive paper chase.

[B: 13:19.9]
A nightmare usually, right?

[A: 13:21.4]
It turned into basically running a simple query thanks to the electronic records.

[B: 13:25.7]
And what's really telling about that story is that this kind of rapid, significant change, it didn't require some superhuman effort or heroes working weekends. And it required configurability coupled with good governance. And crucially, it meant the template they built on that pilot line could be lifted and applied to other lines with relatively minimal tweaking.

[B: 13:44.1]
That's the scalability payoff.

[A: 13:45.8]
Yeah, that's powerful. But configuration alone isn't magic, right? How do you make sure people actually adopt these powerful systems successfully?

[B: 13:54.0]
That's a critical question. Because even the most elegantly configured system will just sit there if the user experience is fighting the operator on the floor. The article makes a great point. Adoption is a design problem.

[A: 14:05.8]
Okay, unpack that. A design problem.

[B: 14:08.0]
It means adoption speeds up when the design happens at the Gemba right there on the shop floor, working directly with the operators and supervisors who will use it every day. It's about designing the interface so the next step is obvious, maybe constraining choices to guide them. It's about instrumenting the work just once, but using that data capture for multiple spc genealogy, automatic audits, performance tracking.

[B: 14:30.6]
And the advice is treat user feedback like a prioritized backlog. Iterate on the design weekly based on what you hear. Training helps, sure, but a good UX and fast feedback loops make it stick.

[A: 14:42.6]
What are some of the common traps? Where do manufacturers still sometimes stumble when trying to implement this?

[B: 14:48.3]
Yeah, there are definitely pitfalls to watch out for. First, and maybe most importantly, remember, customization is not configurability. If every meaningful change you need to make still requires a developer to write custom code, then you haven't really escaped the rigidity problem, you've just rebuilt it.

[B: 15:04.0]
You need to ensure changes are genuinely configuration driven.

[A: 15:07.3]
Makes sense. Don't fool yourself.

[B: 15:08.8]
Exactly. Another huge one is skipping governance. Without versioning approvals, controlled promotion, you will end up with chaos trying to manage dozens of slightly different undocumented workflow versions. It becomes impossible and negates the whole point, right?

[A: 15:24.7]
What else?

[B: 15:25.2]
Avoid going broad before deep. Don't try to implement everything everywhere all at once. The device is solid. Establish a repeatable, successful pattern in one value stream. First prove it works, Learn from it, refine the template, then scale it. Don't boil the ocean.

[A: 15:40.2]
Good advice for any project, really

[B: 15:42.0]
it is. And this one's foundational. Weak master data. If your product definitions, your bills of material, your equipment models are inaccurate or incomplete, it will absolutely sabotage any automation you try to build on top. It really is like building a house on sand. The best MES system can't fix fundamentally bad data feeding into it.

[A: 16:02.6]
Garbage in, garbage out. Still applies.

[B: 16:05.0]
Always. And finally, treating UX as an afterthought. We talked about adoption being a design problem. If the interface is clunky, confusing, or slows people down at their workstation, adoption will stall fast. That freshen point is probably the quickest way to kill momentum.

[B: 16:20.3]
The operator experience has to be front and center.

[A: 16:23.0]
These pitfalls seem avoidable with planning, but definitely need attention. You mentioned earlier where configurable MES really shines.

[B: 16:30.4]
Yeah, it delivers maximum value in those really dynamic environments. High mix manufacturing, engineer-to-order shops, and heavily regulated industries. Think about operations with complex routing, maybe ad hoc steps, needing to select components based on specific attributes, handling parameterized recipes or batch variations, needing early serialization for tracking, or complex scheduling that has to factor in machine setups, operator certifications, tool availability

[A: 16:57.5]
Lots of moving parts.

[B: 16:58.5]
Exactly. All those intricate scenarios benefit hugely from a platform that can be reshaped quickly without waiting months for a development cycle.

[A: 17:06.4]
And of course, you need to measure the results.

[B: 17:08.5]
Absolutely. Like any strategic initiative, you have to measure what matters. The article stresses quantifying success with metrics tied directly to operational performance, flow, quality, stability. So you're tracking things like actual time to value improvements in throughput and yield system stability, user adoption rates, tangible gains in quality and compliance, and critically, scalability.

[B: 17:32.8]
How easily and effectively you can reuse that template for the next line or the next site. It's not just about installing software, it's about driving measurable business improvement.

[A: 17:41.7]
Okay, so as we bring this deep dive toward a close, let's just circle back to that core message Modern manufacturing is complex. It's customized, it's constantly changing. And to really succeed in that environment, plants need an execution layer, their MES that can be reshaped just as quickly as the work itself changes.

[A: 17:58.7]
A configurable composable MES provides that agility. We've heard how it speeds up time to value, lowers ownership costs, scales effectively with governance, and really importantly, creates room for ongoing innovation.

[B: 18:09.2]
That digital thread connectivity is key, right?

[A: 18:12.0]
So the practical advice in the article boiled down this seems pretty clear. Start focused. Template what works. Scale with discipline.

[B: 18:19.9]
That sums it up well. Start small, prove it, then grow it smartly.

[A: 18:24.0]
So as we wrap up, maybe a final thought for our listeners to chew on.

[B: 18:27.3]
The next time you hear about a major supply chain disruption, or maybe a sudden unexpected shift in consumer demand forcing manufacturers to pivot, consider this the difference between crippling chaos and continued progress, between stalling and scaling. It might just come down to how quickly how effectively a company's core operational systems can be reshaped and reconfigured.

[B: 18:48.6]
Makes you wonder what else in our lives, what other complex processes or even industries could could benefit from this kind of adaptable, configurable thinking, building on a stable core, but allowing for rapid change at the edges without constantly having to rebuild everything from scratch. Something to ponder.