Industrial Automation Magazine India

AI/ML also enables better product development and quality improvement

Published on : Tuesday 08-02-2022

Ramnath S Mani, Managing Director, Automation Excellence.

Which forthcoming advances in technology will impact industrial automation?
The convergence of Information Technology and Operations Technology (IT/OT convergence) has been the basis of rapid progress for successful Digital Transformation for the manufacturing industry. This has led to increasing demand for tighter integration and more information that leverages Industrial IoT, Industry 4.0, 5G, Cloud, Edge, Additive Manufacturing, Advanced Analytics, Digital Twin, AR/VR, AI, ML and other emerging technologies. Integration of Power and Automation along with Convergence will lead to getting information about electrical assets and the production process to help improve sustainability across the entire lifecycle of increasing demand for tighter integration and more information.
 
Remote operations will enable Connected Factories, which will greatly benefit from technologies such as Augmented Reality (AR), where the remote user can see any asset in the plant with information digitally overlaid. AR devices sense what the remote worker is looking at and displays the data needed for the operation at hand using tablets, smartphones, smart glasses or wearable computers.
 
Edge Automation platforms designed to provide the full capabilities and benefits of computing at the edge, such as gathering, analysing,  processing and storing data locally in real-time at or close to the point that the data is gathered.
 
The widespread use of Digital technologies and Cloud will necessitate cloud-enabled capabilities and Cybersecurity.
 
How are manufacturing industries leveraging AI/ML? How do automation controllers provide the necessary platform?
Manufacturing industries will leverage AI/ML for intelligent, self-optimising machines that automate production processes. AI/ML not only finds wide acceptance in the manufacturing process but forms a backbone for Smart Maintenance. Complex AI algorithms like neural networks and Machine Learning are a source of trustworthy predictions regarding the status of assets and machinery bringing down unplanned downtime cost in factories.
 
AI/ML also enables better product development and quality improvement. Algorithms that are based on data from results of years of operational experience, examine all possible variations and generate optimal solutions that can be evaluated by pre-trained deep learning models. AI and ML form an essential part of Industry 4.0, and help in improving supply chains, making them interactive to changes on the market place.
 
Automation controllers are the essential source of information and data based on which the algorithms operate. The challenge always lay on the collection and segregation of intelligible information and its connectivity through a common protocol. Technologies at the edge have overcome this challenge. 
 
AI and ML need massive amounts of data to be gathered by IoT devices. What strategies do industry plan to collaborate in data collection?
The matter of collecting data from Edge Devices and legacy systems that include all types of Sensors, PLC, Motors, Drives, etc., and other devices that control the production parameters of the machines that are used in production lines, has always been a challenge for the OT-IT integration.
 
The more common way of getting data out of smart sensors is to use a bridging gateway, which receives data from the devices and sensors, and makes it usable.
 
The process of collecting and processing data locally at the Edge with IoT technology saves storage space for data, processes information faster and meets security challenges. One of the efficient methods of collecting Data and using it effectively is to de-link Devices from Applications. This can be done through a MQTT broker that works in a Publish – Subscribe mode. Any number of Devices and any number of applications can be connected for AI and ML applications without loss of any data in this model.
 
How can AI and ML help companies create predictive models, analyse operations, make accurate forecasts and automate supply chains?
Essentially, AI and ML are tools that help businesses use the enterprise data effectively and through algorithms that collate instances through this data to provide solutions. They curate data quickly for multiple business scenarios and using cognitive predictions provide optimal solutions.
 
Supply Chain challenges are overcome by predicting demand across multiple product segments and multiple geographies. Complex algorithms that tie both the vendor side and customer end with the manufacturing process for increased production and productivity have to be implemented. Identifying trade-offs with hundreds or thousands of interlinked variables and innumerable technical constraints have to be factored into the system to provide an optimal solution.
 
The full potential of AI and ML is realised only when the scale of operations is big enough. How can the average SME benefit with their limited resources? 
The scale of operations has to be big enough to justify the investment in AI and ML as the Capex needed to implement is large. SMEs in India are either dedicated multiple tier vendors to large manufacturers forming a ‘Hub & Spoke’ model or independent direct-to-market vendors.
 
In the former case, it is possible to be a ‘Spoke’ partner to the ‘Hub’ manufacturer and use the Asset of the main manufacturer to form an ecosystem. This will entail a much smaller investment on part of the SME mainly at the Edge while taking advantage of the investment of the main manufacturer. Such a Model can be possible between the manufacturer and a dedicated SME vendor with a long term relationship.
 
In the latter case the only way seems to a Subscription based SaaS model where the Service Provider has the major investment and he is able to provide multiple SME vendors a Subscription based Service for the usage of various functions like Track & Trace, OE and  solutions based on AI and ML technologies.     
 
The human element remains critical in deployment of new technologies. How is skill development to be planned in a scenario of not yet mature technological advances?
This has been a continuous challenge in our country where there is no dearth of engineering talent but not skilled according to the needs of the industry. The rapid changes in technology also does not help the situation. However, there is a tremendous effort by the government and the private sector, through Training Institutes, in re-skilling the resources to match up to the demand. This is a slow process but is making decent progress. There is also a revamp planned in the technical education model that will produce technical resources that will be ready for the future.  
 
Ramnath S Mani is a Serial Entrepreneur and one of the pioneers in India in the field of Industrial Automation including Power Electronics, Variable Speed Drives, PLC, Distributed Controls, Scada, Information Systems and MES. He has been instrumental in introducing Automation in industry in India since early 1970s. Played a pioneering role in bringing to India international Automation companies like CGEE Alsthom of France, Allen Bradley of USA (now Rockwell Automation), Stromberg of Finland (now part of ABB), Exide Electronics of USA and Control Techniques Emerson Industrial Automation of UK and USA., Inductive Automation of USA, among others. Having a deep domain knowledge on the application of Automation in the manufacturing sector like Paper, Steel, Cement, Sugar, Textiles, Rubber & Plastics, Automobile, Material Handling, etc., Mr Mani has considerable experience in implementing the transition of Industrial Automation from Analog to Digital Technology.