Contributed by: Raghavendra Singh
I work as a production control engineer at a PSU under the Ministry of Defense which primarily deals with Electronics equipment for Defense forces. Indian Army, Navy and Airforce, and other departments of India and overseas are our customers. I work in the product support department, where we have Annual Maintenance Contracts for Various RADAR systems deployed at various sites throughout the country.
Our RADAR systems need to operate with 99.9 % serviceability as it’s one of the important parts of defense systems that make our job very critical. My responsibility is to plan in advance and be ready with items, and production of various critical assemblies may be required to keep the RADAR working before any actual failure. In 2019, when I joined the department, production and procurement were going on pace as per anticipated requirements, but there was no proper maintenance of dynamic production and procurement status, which led to focusing on mainly critical cases. And every day, there was a pop-up of a certain critical assembly or procurement.
Working with SAP technology, we have a lot of data to study, and by analyzing that data, we can improve process, production, procurement, save time and minimize cost also. We can automate our progress monitoring also.
I told my superior that I am doing a Data Science course, and they asked me if this knowledge can help us with making a better status and keep the production targets in line to meet the requirements before they become critical.
Then we started our journey of improvement. We collected all the procurement and production data from SAP and requirements records for all RADARS systems under AMC and organized all the useful variables like procurement time, failure rate, cost, vendor data, delivery date, mode of procurement, etc. We analyzed the whole data and using descriptive statistical tools and came up with a relation between failure rate and procurement. We built a predictive model using regression tools and improved our procurement and production process by 70%. We worked on different parameters and built a status for keeping track of the project, including various statistical and visualization tools.
First, we tested our predictive model for procurement planning on a small project. We trained it for the past seven years and tested it for the last two years. And interestingly, it had an accuracy score of 86%. Then we implemented it on various other projects also. Our Status file is doing extremely well in improving our process and keeping track of procurement and production activities. Earlier, we were giving most of our time in handling critical cases only; using status files, we are able to reduce our critical cases by almost 70%.
Production/procurement status is helping us in monitoring and effective control of the projects and has a significant effect on closing customer complaints in minimum time.
By our model, we can predict the requirement and production issues before time and start the procurement and preventive action before time. Also, taking the required action on time using our model and improving the overall performance of the project has improved customer trust in us and added value to the department. For me, it’s a great opportunity to utilize my knowledge and gain valuable experience. I am still working on it for more improvements.
“Torture the Data and It will confess to anything.”- data is the tool these days that can be utilized to achieve significant results.
Want to learn such concepts? Join Great Learning’s PGP Data Science and Business Analytics Course and upskill today! For more such success stories, watch this space.
