A lot has been written, said and discussed in the domain of Artificial Intelligence. From the Turing test conducted by Alan Turing in 1950 which offered an opportunity to understand whether machines can exhibit intelligent behavior to AutoML (Auto machine learning) by google which claims to reduce the dependency on humans to build AI models, the technology has come a long way. However, the question that still intrigues many is whether this new wave of digital intelligence is intelligent enough to create value. This is one of the biggest challenges C-level executives in the manufacturing industry face when they propagate the idea of investing in this technology. Preparing a business case and binding the investment to the RoI, in an asset-heavy industry, becomes a daunting task and many at times hinder the buy-in or progress of such programs across the manufacturing enterprise. The risk of failure is perceived to be large despite the inherent advantages of AI technology and most of the companies try to reduce the risk by shifting the ownership of assessment and implementation across plants to technology consultants, vendors and value chain partners.
Artificial Intelligence solutions are being increasingly deployed across sectors such as healthcare, logistics, telecom, education, fintech, banking, ecommerce, agriculture, entertainment industry, sports, media etc. and these sectors have realised the benefits too. Manufacturing industry has been the forerunner in adaptation of technology and is expected to derive the maximum value from the implementation of technologies. So whether it is the use of Industrial IoT (IIoT) which offer humungous data on machines and their performance and paves ways for predictive maintenance, or the use of big data along with mathematical models to predict the demand forecast, a lot can be offered by AI to transform the industry.
The use of AI is also being propelled by GoI and the National Strategy on Artificial intelligence mandated in FY2018-19 with the intent to create an ecosystem of AI technology became the stepping stone in this direction. As per a report by Accenture AI has a potential to add approximately 1 trillion to the economy of India by 2035. This gives the necessary push to manufacturing enterprises to identify and implement opportunities in the field to offer technological advanced products and services, creating customer stickiness.
However, the question remains – What RoI can those automotive enterprises which invest or intend to invest in these technologies generate? One way to look at the return on investment is the improvement in productivity across the manufacturing value chain, the valuable business insights the technology generates such as manpower productivity, process performance and agility, and asset utilization, to take necessary business decisions. One advantage of the digital economy is the availability of a huge amount of resources which reduces the technology adaptation and learning curve for organizations which are part of the manufacturing sector, creating a low barrier to entry. This leads to ease of product imitation and extensive competition among the participants for the share of business, decreasing the value delivered to the organization and its customers. Moreover, as the volatility and complexity of technology increases from one model to another, i.e. as data velocity, veracity, and volume increases, and as models adapt and self-optimize to these changes and graduate from basic search algorithms to complex neural networks and to machines teaching machines, enterprises can no longer create value by offering technologically advanced products alone.
In an asset-heavy manufacturing sector, strategic advantage must be created by evaluating areas that cannot be easily imitated, areas such as processes. It is the unique attribute of the non-imitability of the process that gives a competitive advantage to many manufacturing enterprises across the world and ensures the loyalty of the customers. It is this area that has remained largely unimplemented across manufacturing setups despite it being the forerunner in the area of automation and technology. IoT devices and sensors can collect data on existing manufacturing process parameters which can be analyzed by a neural network to identify NVA (non-value added) activities, opening up opportunities to redesign the process, convert existing setup to lean and creating unique opportunities for the enterprise. With the shift in the customer expectations from a functional product to a low cost, feature-rich, high-quality vehicles, the price pressure has increased on the auto manufacturers who are already struggling with low demand due to Covid–19, high asset base, huge inventory carrying costs, and skewed fixed costs. Redesigning processes by use of the insights generated by implementing AI, these auto manufacturers can create a great strategic advantage for their customers through optimization of costs.
Automotive players can use the McFarlan’s strategic grid (Fig. 1) to map the present system and to plan for technologies they would want to adapt in order to generate competitive advantage and to align their operations to the business strategy. The focus should be to adapt those projects which can help the enterprise to move from the ‘Factory’ quadrant to the ‘Strategic’ quadrant.
Business case for AI adaptation hence can be strengthened by showcasing the alignment of the new technology initiatives to the business strategy. Process improvement projects in manufacturing setups can include the study of man and material movement through use of legacy CCTV infrastructure to earmark existing routes and the AI solution can be used to optimize these paths with the aim to optimize the cost function. Impact on bottom-line can be assessed by conducting a study on the manpower deployed across the factory operations and efficiency improvement opportunities can be predicted by the AI-enabled systems. High maintenance machines and equipment’s along with NVAs can be identified to isolate problems and to further the improvement of the process. Non-imitable process redesign to create strategic advantage should become the ultimate goal for all the players in the automotive sector.
Though auto sector has always been an early adopter of technologies which is evident from its foray into EV segment and same is true in case of AI technologies too, there still remain many opportunities for implementation which can be aligned to the overall business strategy. Delay in undertaking critical AI projects can result in loss of competitive positioning, creating a huge barrier to growth of market share and the bottom line especially when the auto sector is already struggling with keeping the momentum of the cash flows and Covid–19 is expected to create further degrowth.
AI technology is here to stay and the best time to unleash the might of these technologies in the automotive sector is now.
Pawan Chhibba is an independent consultant to organizations in the area of Artificial intelligence and Machine Learning. An IIM Calcutta Alumnus, Pawan has more than 15 years of rich experience of working with C-suite executives in MNC and Start ups in India and USA. His interest areas are Digital manufacturing, IIoT, Data analytics, ML & AI.
Published in Telematics Wire