What About Data Management?

Davina McDonnell

By Davina McDonnell, Global Marketing Manager, Cogiscan

Connectivity is the mandatory foundation for ANY factory digitalization initiative. And while connectivity gets a lot of hype, the less talked about challenge is how to manage the vast amounts of collected factory data. So, while the electronics manufacturing industry is more than six years into making factories smarter, many manufacturers still struggle with both collecting and managing factory-wide data.

Connecting & collecting data

Most electronics manufacturers have a diverse equipment installed base consisting of a wide range of vendors, models, and vintages. Typically, each of those vendors (and even vintages within vendors) have different requirements to tap into the data source. It’s no easy feat to access these critical data points, and typically requires extensive collaboration with the many different equipment manufacturers.

Rather than wasting internal resources to try and understand all the nuances and requirements of the different machine types and vintages, it’s a better choice to rely on out-of-the-box adapters to immediately tap into production data.

Many manufacturers, including Jabil, turn to Cogiscan’s Co-NECT platform to efficiently and properly connect and access all their machine data.With the largest library of machine and software interfaces, Cogiscan offers unparalleled access to the best and most complete data sources for any SMT equipment.

“After our internal evaluation demonstrated that Cogiscan’s Co-NECT technology would allow us to connect the majority of our SMT equipment within a few months, they became our top choice to accelerate our digital transformation and enable us to move towards a more predictable environment. Additionally, the Co-NECT platform offers the richest data among the alternatives that we evaluated,” Colin Reid, Senior Director of Digital Factory, Jabil explained.

Massaging & managing data

If only accessing the data was enough – exported raw from the floor, data always needs to be filtered. To take intelligent action, factory-level data should be consolidated within a central storehouse, enriched with domain appropriate contextualization, and shared through an open architecture platform.

Cogiscan’s 24+ year history of SMT specific domain-expertise ensures that collected data is contextualized appropriately – that it gets enriched to be usable. Providing meaning to your collected factory data, this guarantees you can properly analyze manufacturing metrics and KPIs specific to your needs.

This step is critical as it establishes the “Truth” for factory digitalization efforts – and the systems that drive actions and make decisions, including the traditional MES/MOM, and ERP tools, must be fed accurately.

Structuring data to feed AI tools

Now that AI is a major focus for a lot of manufacturers and equipment vendors, everyone is scrambling to figure out how to use collected factory data to feed these intelligent systems. Most electronics manufacturers are sitting on troves of data and are looking to maximize that untapped potential to better optimize manufacturing efficiency. As with all things related to connectivity and data in this industry, it isn’t as simple as feeding the AI tool data as-is, the structure and format of the data is paramount.

Collected factory-wide data needs to be structured and formatted in a way that the algorithm knows how to access it, read it, and interpret it – essentially, so the AI platform knows what to expect every time. These intelligent systems are only effective if the information feeding them is meaningful. To avoid another “Garbage in, garbage out” scenario, consider how all collected factory data will need to be structured to feed the machine-learning platform.

With the right data management system in place manufacturers can take advantage of these powerful and intelligent applications for the shopfloor, using AI-enabled advanced analytics to detect and prevent problems before they even occur. At Cogiscan, we have a few early adopters of our AI-based IIoT.Edge platform using specific SMT use-cases to solve every day manufacturing challenges – such as reducing false calls, correlating printer data to SPI results, preventing unplanned machine downtime, as well as leveraging SPC to eliminate waste.

These rapidly evolving technologies will continue to transform how electronics manufacturers operate – ensure your connectivity and data management strategy is ready, so you too can unleash the power of AI on your shop floor.