Why model-based transformation in manufacturing is essential to replace ageing IT and spreadsheets

Adrian Wood explains how disconnected tools create translation loss between engineering and the shop floor, turning late engineering changes into avoidable cost, scrap, and schedule risk. he goes into more depth on: what transformation looks like with a single source of truth connecting product, process, resources, and constraints; what changes operationally with impact analysis in a virtual twin before execution; constraint-aware scheduling with what-if scenarios; and why manufacturers should act now, as complexity and speed have outgrown spreadsheets, with AI-assisted decision support capable of helping planners

How often have you had that sinking feeling when seeing a late engineering change order? Do you relive the stress of the cascading events? Someone updates a spreadsheet schedule, someone else updates an MBOM extract, a supplier works off an old revision, and the shop floor discovers a feasibility issue during ramp.

Spreadsheets and legacy systems are not inherently ‘bad tools’, they were just not designed to support the complexity and speed of today’s manufacturing environments. These tools might be able to scale information within their own scope, but cannot provide the digital thread that orchestrates information across silos, and revels the true impacts of changes.

The overall business impact of inaction includes slower time-to-market, lower productivity, and higher costs related to quality and scrap. This, of course, also means there is a significant opportunity and ROI to transformation!

What does transformation look like?

Do not think about transformation as just new software, it is really a shift to a process of model-based governance that uses a unified data foundation, and empowers the workforce to make better decisions and drive innovation at greater scale.

Today, your teams are probably spending significant amounts of effort translating and reconciling: engineering hands off documents, manufacturing rebuilds definitions, planners re-enter constraints, and the shop floor finds problems late. In the transformed world, the enterprise works from a shared, living definition, so change becomes faster, safer, and repeatable.

Here are some of the key elements to consider when thinking about the best-practices of transformation:

• A Unified Data Foundation – transformation accelerates when teams start to use (and trust) a single source of truth. Therefore, a common data model is key to connecting critical information across product definition, process plans, resources and constraints. This allows seamless versioning, traceability and the knowledge that the ‘current state’ is always known.

• Orchestration Between Silos – eliminating ‘translation loss’ between engineering and manufacturing is the biggest ‘leap’ in the transformation. Being able to rapidly see the impact of changes eliminates costly, real-world corrections (seeing the impact on lines, tooling, suppliers, work instructions, etc.). Note that this link is bi-directional, so that governance works whether it starts upstream or downstream.

• Virtual Twin Simulation – feasibility needs to be proven before you commit to execution. This can only be efficiently accomplished in the virtual model of the production systems. Virtual twin technology leverages the unified data model, and permits the governance for team to experiment and visually analyze impacts and innovation in a scientific and collaborative environment.

• AI-Assisted Decision Support – developing ‘optimal’ plans requires humans to let the algorithms do the heavy lifting. The complexity of manufacturing operations leads to an almost infinite array of possibilities based upon real-world constraints and changes to demand and supply. AI has become adept at consuming vast amounts of real-time (and historic) data to help guide planners to making better-decisions that are feasible and trusted.

Although there are more aspects to a complete transformation, high tech manufacturers are leveraging these foundational elements to move away from manual methods and disconnected tools. Being able to align around trusted data, simulate and make optimised decisions, allows them to consistently respond faster, and with less risk.

What are the results?

Leaders care about impacting the core elements of their products and business: speed, cost, quality, service and risk. There are many daily inefficiencies that are the root causes of poor business performance, but digital transformation can address all of these with significant results. Here are some operational ‘before and after’ examples that are typical:

Existing Challenge ‘After’ Transformation

ECO meetings debating whose file is correct Instant impact-analysis in the virtual twin

Ramp issues discovered in pilot builds Feasibility proven as part of design/engineering

Schedulers chasing parts in spreadsheets Constraint-aware schedules with what-if scenarios

Training via tribal knowledge Visual, validated work instructions (and ergonomic checks)

Quality investigations via email threads Traceability to exact revision, lot, and operation

There are, certainly, many more examples of root-cause challenges, but the broad scope of virtual twins allows companies to model and simulate almost any part of the process needed, and the scientific accuracy is able to capture any level of detail.

How to get started

A good place to start is usually where the biggest challenge or impact occurs, although it is feasible to run smaller pilots (especially if the company is new to transformation, AI technology, or perhaps has a ‘shaky’ data foundation). Regardless, choosing a single value stream at a time is appropriate; that could be NPI ramp, ECO responses, quality escapes, etc.). This ensures that there will be defined goals and stakeholders to drive the process.

Next, companies should focus on building the core backbone of the model (including data) to define the product, processes, resource and constraints). Developing this virtual twin is key, because making confident and trusted decisions relies upon a solid foundation.

After these steps, companies can start to define and experiment with the connection points and governance between engineering and manufacturing. This stage provides the first steps into the ‘playground’ ‘that will become the centre of the transformation, allowing stakeholders to not only align on common models and decisions, but now to experiment with new ideas and opportunities for innovation.

Finally, as each decision is made, the virtual twin can validate and record the outcomes and provide a closed loop from execution back to engineering that drives continuous improvements.

What is the value?

Of course, it depends upon the problem that you are solving, but the important take-away is that the transformation of day-to-day operations does impact all of the critical business goals that leaders will be expecting.

• Faster ramp / fewer late changes
• Lower scrap/rework and fewer quality escapes
• Better schedule adherence and service levels
• Lower inventory buffers and less obsolescence
• Higher engineering productivity and repeatability across sites

It is important not to forget the value to the human element of transformation. There are many warnings that digital transformation and AI will reduce and eliminate the manufacturing workforce, but they are not completely justified. Firstly, most manufacturers are still struggling with missing gaps in workers and skills, so some of the transformation will help to fill the empty space, but existing workers will also benefit from becoming far more productive and consistent (even with turnover).

For further evidence, Panasonic Connect is an example of a company that has undergone digital transformation to great effect. Originally facing problems of static spreadsheets, a lack real-time collaboration, and the challenge of the ‘data drift’ between engineering and the shop floor, Panasonic Connect looked to DELMIA solutions for help.

DELMIA allowed Panasonic Connect to move away from analogue and fragmented data to a unified platform. By digitising their expertise and connecting it to a ‘Single Source of Truth’, they were able to visualise the entire factory in real-time. This transformation allowed them to track progress accurately and synchronise global operations; a direct result of replacing ‘Ageing IT’ with the 3DEXPERIENCE platform.

We could centralise and manage our unorganised data and analogue info and because of that we could track in real-time the overall factory line operations.

Spreadsheets and ageing IT really cannot keep up with high-tech manufacturing complexity. This means changes propagate slowly, issues surface late, and teams waste time reconciling data instead of improving performance. If you would like to learn more about this topic, you can explore this exclusive whitepaper.

Adrian Wood is Director of Strategic Business Development at DELMIA.