Executive Summary : | The alloy design is an open field that depends on many factors such as processing conditions, availability of resources, properties required, etc. Recently the design of refractory high entropy alloys (RHEAs) is getting wider attention due to their applicability in high-temperature conditions. Identifying the RHEA compositions with reasonable room temperature ductility and high-temperature properties is still challenging. The design of materials can be accelerated by applying machine learning algorithms. Several ML algorithms are available in the literature, and selecting a suitable one for a particular data set required proper testing of each algorithm. Identifying proper material descriptors is important to improve the accuracy of prediction. The efficient utilisation of feature selection algorithms can be explored for this purpose. The current study focuses on developing a materials informatics workstation with trained ML models, feature selection algorithms and data visualisation. The workstation will be able to predict the suitable material descriptors and algorithms for a particular application. We will target developing the RHEAs with enhanced hardness and ductility using this workstation. The workstation development includes data generation from experiments and literature, analysis and ML algorithm development and testing and validation using experimental data. Significance of the currently proposed research • Development of Machine Learning algorithm for RHEA design • Application of feature selection algorithm for identifying significant material descriptors • Identify the important design parameters for RHEA design • Identify RHEAs composition which desired phases and properties using trained ML models |