By: ICN Bureau
Last updated : April 04, 2025 8:23 pm
This cutting-edge solution is completely made in-house with domain specific knowledge of the process industry
Jaaji Technologies, a well-known player in the digital solutions space for manufacturing industries, has launched its new age solution inSis.ai (industrial smart information system) targeted at resolving use cases that are emerging across industry verticals such as oil & gas, chemicals, pulp & fibre, steel, cement, paper etc.
The solution is built on its already proven inSis platform which is implemented in 100+ sites across 75 clients. Some of the top use cases new software is focusing on are: Predict asset anomalies and failures; Real-time batch deviation monitoring; AI based optimization guidelines; Predict in-process and finished goods quality parameters; Summarization of operator comments; and AI Assistant for referring process information from historical data.
Jaaji Technologies acknowledges that this is a multi-year journey the clients have to embark on from the step of data consolidation to aggregation and later to analytics. With years of experience under its belt, the team is rightly positioned to partner with the manufacturing clients to bring in the intelligent layer of analytics into the transformation programs.
This cutting-edge solution is completely made in-house with domain specific knowledge of the process industry.
K. Siva Rama Brahmam, CEO, Jaaji Technologies said, "This is a unique value proposition which is a blend of chemical engineering principles and software design philosophy the team adopts. This platform is quick to be deployed, and users can plug-in their custom-built AI models to predict the parameters of their choice. In addition, the Jaaji team also has developed some in-house models which are ready for commercial deployment at the clients. Its strength lies in its versatility to be deployed across different verticals with low cost of ownership.
As expected, inSis.ai deploys sophisticated Machine Learning Algorithms like Advanced Pattern Recognition (APR), Artificial Neural Networks (ANN), Multi Variate Regression (MVR) etc. to study the historical data set and predict the output parameters.
inSis.ai has 3 sub modules serving different objectives: PredIT - predicts process and equipment anomalies, quality parameters using deep learning models; OptimizeIT – identifies relationships between process variables and suggests the optimal combination for best output; and AskIT – Uses NLP (Natural Language Programming) to summarize information and contextualize into actionable insights.
Talking about captive power plants, Siva said, "The team used power factor data across 150 plus equipment to identify anomalies in power consumption for different loads and suggest potential root causes and remedial actions. This directly led to savings of about Rs. 50,000 per day for the plant. In the quality area where the real-time quality parameters of a manufacturing process can be predicted based on the lab analysis results that are shared with the operational teams. These types of analysis not only save time but also directly impact the operational bottom line justifying the investments made."
"For a refinery, the team delivered a synopsis of the observations entered in the digital logbooks highlighting the recurring incidents, anomalies and safety related comments. This helped the maintenance in-charge to schedule the maintenance activities and avoid potential downtime of the critical equipment," added Siva.
The possibilities are endless as manufacturing organizations have gigabytes and terabytes of data from years of operations. It is up to the leadership to invest in the relevant use cases and derive value by improving the operations, saving energy bills, ensuring safety of the personnel and the plant.