Infosys TechCompass

High-tech

Trend 3. In-factory and supply chain operations reach end-to-end digitization

Industry players strive to create a digitally integrated system that collects data on all objects and processes within the factory and on external supply chain elements. Such end-to-end digitization of materials, machines, and processes enable faster changeovers, single-minute exchange of dies, and plug-andproduce capabilities, resulting in improved yields. It is achieved through a comprehensive network of sensors and IoT devices, which are managed through a central supply chain management system.

This kind of system also enables dynamic machine allocation by informing factory workers about the readiness of machines. For instance, it tells which machines are lying idle, which ones are nearing completion of the previous batch, and which ones are ready with the required dies already fitted. This helps optimally choose the correct options for maximum yield and minimum loss of resources (including time).

Further, such connected systems help form a digital thread across the supply chain, which provides complete visibility into operations and schedules of suppliers, warehouses, and internal processes. This synchronizes production plans with shipment timings, and order picking and packing with transportation schedules of third-party logistics partners. AI/ ML augment these systems to make dynamic predictions for demand, supply, resource availability, and utilization rates, based on evolving economic conditions and other factors.

Infosys engaged with a leading enterprise networking and security equipment manufacturer to build an integrated digital supply chain solution to help the firm gain complete visibility into its operations. The manufacturer invested approximately $2 million in this initiative, and was able to improve productivity and optimize order fulfillment by connecting its supply, demand, and inventory management operations. It surpassed its ship-to-first commit target of 95% with improved on-time delivery performance while reducing lead times for inbound supplies. Essentially, in addition to the production yield, the company improved its customer satisfaction scores too.

High-tech

Trend 4. AI-based predictive maintenance and smart automation

AI-based predictive maintenance solutions prevent equipment/machine shutdowns and efficiency losses by proactively attending to maintenance issues and ensuring the timely availability of spare parts. Data collected from various equipment/machines in real time is used to detect patterns and improve maintenance schedules, which leads to better firsttime fix rate. It also helps predict the working age and utilization rates of equipment and parts, which can even be benchmarked against industry's top quartile performance. Moreover, such smart systems suggest corrective actions to keep the machinery running at maximum efficiency, and alert designated users if required.

Additionally, AI-based digital supply chain solutions help integrate automation across various nodes of a supply chain. For instance, warehouse automation allows just-in-time practices within factory operations, and enables automatic picking of materials from storage facilities. Further, RPA reduces human intervention across supply chain processes, creating time to identify and undertake further improvements. Also, reduced human errors improve wafer processing, leading to improved yield rate.

Such solutions are further complemented by digital twins or virtual replicas of the entire supply chain and operations. When data is delivered on digital twins, manufacturers determine the performance of assets and processes through a centralized system in real time and take preventive actions. These solutions help identify and plug the smallest of productivity and quality gaps, improving yield.

A leading U.S.-based electronic components distributor wanted to improve efficiencies in its order management processes through automation. Infosys deployed an AI-based automation solution to remove manual interventions from tasks such as prediction of the block removal date for a backlog sale order. It also alerted account managers and triggered ERP systems to remove the block. This significantly reduced staff hours, allowing more time for further process improvements.