When one speaks of industrial IoT, the first subject that comes to mind is sensors.
Location. Equipment status. Pressure.
Production data.
Temperature.
However, in a complex industrial ecosystem, other issues take center stage like "What is actually taking place surrounding this data?":
A work order may exist in a database or a machine might transmit its status. However, operations teams would be better off by knowing the actual physical whereabouts of machines, people, goods and other resources.
One such area where Artificial Intelligence and IoT could jointly work wonders is this one.
Moving on from IoT data to real context
With IoTs available on assets connected, one can gain an insight into these assets. However, Artificial intelligence takes this to the next level by analyzing patterns existing within this information.
Imagine a food processing plant. Ingredients traverse through receiving, storing, preparing, packing, cooking and so on. Meanwhile the human operators and reusable packaging systems flow too.
The movements and whereabouts of these resources can be mapped.
It is possible for artificial intelligence to analyze the established patterns and identify bottlenecks and workarounds. The object is not to gather raw information but to establish context.
An industrial case of changeover of manufacturing process
This is best illustrated by beverage manufacturing. Production units often need to be switched between products manufactured. When changing a product line from one flavor to another, not only would tooling and containers need to be switched over, but so would bottles, caps, labels and eventually the workforce!
AIoT would enable to coordinate everything on these lines and check resource availability relative to the planned sequence of activities.
This question then shifts from "Why did the manufacturing process take so long?" to "Which of the resources and coordination steps was the problem?"
Physical movement of packaging in the plant
Another area illustrating the importance of physical space is that of packaging and packaging material transport to and from the production floor. Bottling lines or carton lines, require a regular flow of packaging elements like cartons, labels, bottles/cans, caps, tooling, change parts. Analysis of their actual movement patterns in relation to the demand from the production floor would reveal delays in delivery of material, surplus material stockpiling, the movement of operators and ultimately uncover issues related to the physical logistics of operations inside a plant. The problem at hand here is not one of inventories, but rather 'flow' logistics.
Interlocking maintenance records with the actual workplace
Consider water and wastewater operation. These operations may be characterized by thousands of remote pumps and valves, complex electrical controls systems, mobile crews and maintenance shops and even a geographically dispersed pipeline infrastructure coupled with remote pumping stations. CMMS and EAM tools can help in the arrangement of these maintenance jobs and maintain the records for each component.
These records cannot necessarily capture all that is happening out in the field.
AIoT can help correlate the movement of field elements and operators with recorded maintenance tasks and history of a given equipment, adding to efficiency of the operations on the ground and that of resource allocation.
The material-flow problem of waste management as well!
Waste handling facilities face ever-changing material profiles, mobile and immobile equipment, workers, and transport machinery. These would generally move about the plant floor transporting materials from point to point throughout receiving, classification, processing, baling and outbound shipping. Bottlenecks and idle times will easily occur which can be detected through the analysis of the flow patterns by artificial intelligence and by using a computer vision to determine kind of materials if it's a sorted output and determine what to process using intelligence and by also considering factors like actual position and direction of moving object.
Knowing "where it is and in what direction it is going,", rather than just "what material type is inside container?"
brings new dimensions of intelligence into operations.
Building the framework beyond technology
For an intelligent operation using both IoT and Artificial Intelligence to become successful, one should think of "Which decision is becoming difficult due to absence of context from the field" not "where to implant artificial intelligence?" This may pertain to finding a component, orchestrating a changeover of manufacturing, delivering the appropriate component to an assembly line, identifying a misplaced field item or batch investigating a particular batch that is in process at the moment, or understanding what process is working to capacity/not working at waste stream. Once the nature of the operation problem is identified, appropriate data from IoT can be used and patterns analyzed using AI.
a widespread commonality across industries
All these are very diverse operational fields like food processing, beverage manufacturing, packing and shipping, water and waste water processes are somehow alike as there is an immense element of constant physical movement of resources throughout the various complex workflows. The location as well as movement details of all resources in such a setup adds valuable operational information to the overall process. Grouping of all Food, Liquid, & Environmental related operations for applications under the roof of Artificial Intelligence and Internet of Things is one such area that is actively being probed and explored in Aperture Venture Studio-covering aspects of fluid manufacturing processes as well as that concerning waste, recycling and water based industries. It would simply indicate to any industrial organization that the effectiveness of Artificial intelligence is greatly multiplied through its integration into the context within which operation decisions are put into practice-and thatcontextis often very physical!
For More Info Visit: apertureventurestudio .com
Top comments (0)