Embracing AI In Today's Industries
As manufacturers around the world make strides to build out the vision of Industry 4.0, many – if not most – will embrace advanced machines to accelerate the digital transformation of their operations. The pressures remain: improve productivity, meet consumers’ expectations for customization and drive continuous product innovation while consistently lowering costs. Bottom line: manufacturers can’t ignore the thrumming beat calling for measurable and meaningful progress on Industry 4.0 in 2018.
It’s not an easy journey. In last month’s post, we called out some of the realities manufacturers face when tacking the wholesale change called for in the digitization of manufacturing. Those obstacles will not easily be overcome, and yet, there’s not a lot of patience for sitting on the sidelines – progress must be made. So, stepping onto the factory floor, what are some the biggest challenges in the transformation from analog to digital operations? In general, these are the most common
- complex, legacy systems, many with proprietary applications,
- huge volumes of data generated by existing equipment and
- few resources to dedicate to deriving insight from the huge volumes of data.
It’s no wonder then that for many manufacturers, the first thing that comes to mind when thinking about connecting machines on the production floor to one another, and then ultimately to back-office systems in finance, purchasing and more is that in no short timeframe, they’ll be drowning in data without a life preserver in sight. Changing that mindset requires a model that’s built on a new paradigm for automation: one that taps the power of software to orchestrate the actions needed.
Changing the Real-world of Manufacturing with Gaming Technology
In the earliest days, automation programming was hard-wired, literally. Programmable logic controllers (PLCs) were the brains behind industrial automation, a technology that relied on cables, relay racks and highly skilled programmers using ladder logic and structured text to direct the machines. The complexity and rigidity made these solutions costly to implement and to change when change was necessary.
No more. Behavior tree technology, used in video games to define basic behaviors for objects, characters, and environments which come together to bring an interactive world to life, blends logical flows and actions seamlessly to direct robots to perform tasks. It’s all done with onboard software – and can be done by the operator who knows best what the workflow is for a task.
While behavior trees make it much easier for manufacturers to put more robots to work on tasks, this approach to enabling automation does so very much more. Starting with the behavior-based robot, manufacturers can build behavior-based work cells, then behavior-based factories. They can create intelligent portions of a line first, transition to an entire line, a collection of lines, a whole factory and then a collection of factories.
With performance, task data collection and introspection built into the design – not as an add-on or layer that sits on top of the design – manufacturers no longer will find themselves drowning in a sea of data. Instead companies will have a powerful way of making sense of the data, and more importantly, deriving value from that data. In the race to build the digital factory, manufacturers will find robots the perfect physical and cognitive partner . These machines make themselves indispensable with the ability to perform tasks, collect and analyze data on productivity, quality, and reliability and cost for themselves and other equipment in the process and make the insights available for interpretation and action. Where do you see the greatest potential for robots to contribute to the brave new world of digital manufacturing?