(Wiseman Innovations- USA, https://wisemaninnovations.com/)
Predict the total cost incurred in a month by an Accountable Care Organization (ACO) in the USA using partial available claims information using time forecasting algorithms like LSTM.
(Grunenthal- Germany, https://www.grunenthal.de/)
Transformed data into MySQL and apply multiple AI/ML models like Regression, Clustering, Summarize Text and Word2Vec for quantitative and qualitative analysis on pharmaceutical data to better understand the needs of targets.
(Robert Bosch - Germany, https://www.bosch.de/)
Build a toolchain to automate model-based risk assessment (MBRA) in parallel with model-based systems engineering (MBSE) using System Modelling Language SysML.
(Integ Consulting - USA, https://www.integconsulting.com/)
Created a scalable architecture using restful FASTAPI server and Spark to perform dynamic ETL on big data workloads. Developed an end-to-end AI/ML pipeline using AWS SageMaker. Developed an AI/ML infrastructure using Spark and AWS services
(Integrated Medical Sensors- USA, https://integratedmedicalsensors.com/)
Build Markov model of a glucose sensor that is to be embedded in humans. Formally verified its response time by calculating steady state probabilities of Markov chains.
(Asystem- France, https://www.asysystem.com/en/)
Ranking of critical components in an Ultra HVDC transmission system, having multiple redundant components like transformers, smoothing reactors, etc., is obtained after one year operation through dynamic fault trees (DFT) analysis using our tool SAFEST. System DFT Stats - Static Gates: 157 (142 OR, 15 AND), Dynamic Gates: 64 (56 SPARE, 8 PAND), Basic Events: 252 Ranking is done based on the following importance measures: Birnbaum Index (BI), Criticality Importance (CI), Risk Achievement Worth (RAW), Risk Reduction Worth (RRW), Diagnostics Importance (DI)
(Asystem- France, https://www.asysystem.com/en/)
Markov model was built for an renewable energy system comprising wind farms and an HVDC transmission system (with onshore and offshore stations). The system was studied for expected repair time, long-run availability, expected energy loss during system outage, expected monetary loss, etc.
(NRTC- Pakistan, http://www.nrtc.com.pk/)
A graphical link analysis solution was built for tracking individuals based on their call record data, criminal records, hotel stays, FIRs, etc. The tool can handle huge amount of data at the same time.