Thunderstorm is a mesoscale system characterized by heavy rain showers, lightning, thunder, hail- storms. In the North-Eastern region of India during pre-monsoon season is affected by severe thunderstorms called ‘Norwesters’ and locally called ‘Kal Baishaki’. To understand the severe thunderstorm characteristics, we have successfully conducted three field campaigns during the period from 15th April to 31st March for the consecutive years 2009, 2010 and 2011 by deploying Micro Rain Radar, PARSIVEL (Particle size and Velocity) Disdrometer, Lightning Sensors, and Automatic Weather Station at Regional Meteorological Centre, India Meteorological Department, Guwahati. The intention of this paper is to develop a methodology for classification of MPS into hailstorm, thunderstorm, convective/Non-thunderstorm and stratiform precipitating clouds on the basis of the thermo-dynamical and microphysical parameters. The algorithm assimilates more than one information bases from Radiosonde, lightning sensor, PSD, MRR and AWS observations. The proposed methodology is evolved on thermodynamical parameters, lightning strokes and length, microphysical parmeters on ground and aloft. This algorithm performance was evaluated by using STORM three intensive observational periods i.e., IOP-2009, IOP-2010 and IOP-2011, respectively.