BI & Analytics Case Study Using AI to Optimize IoT Data in Quality Control for a Global Heavy Vehicle Manufacturer Quality control is a critical function in automobile manufacturing. More so when you are manufacturing premium heavy automobiles like buses and trucks for a global market. Like most digitally advanced manufacturers, our client invests significantly in next-gen technologies like robotics and IoT to minimize faults in production line. But with the sensors and connected devices spitting out complex and huge volumes of data, quality control becomes a data and analytics problem in no time. Challenges: Client sought a proactive & quick system to avoid the scrap in production line Speed up the current manual inspection process in QC Streamline data collection process as the data volume is huge Provide operating limits for the newly identified parameters that define the quality Identify the pattern of more than one parameter causing the faulty crankshaft Our Solution: Leveraging our experience in implementing AI for manufacturing Quality Control, we came up with the right solution. Set up the environment in the Microsoft Azure with additional components of Azure ML, Blob storage and Data lake Created natural threshold to monitor abnormality Built an early warning predictive model Sourced data from IoT devices fitted on the machines in production line Business Benefits: 25% reduction in the scrap rate achieved through the AI/ML Patterns The scalable ML model was able to save the Inspection time and effort by more than 60% It also brought about a 70% reduction in the fault rates Made the inspection process smoother through automatic alerts if parameters values crossing the limits