• Mr. Ganapathy Subramaniam B
• Dr. T. Ramaprabha

Industrial incidents, which cannot be avoided in chemical and gas industries due to various factors involving operational errors, safety measures, defective workings etc. Occupational Health and Safety (OH & S) of India specifies the key regulations to make every place as an accident-free industrial environment. Recording and analyzing the historical industrial incidents data helps the industries to identify the potential trends and helps to reduce the loss. Cause of an industrial incident plays a major role to classify the incident to measure the severity and area of cautiousness. But in most of the industrial incidents, the cause of the incident occurred is unknown due to various factors. Modern Machine Learning techniqueshelp in this case to predict the cause of incident occurrences. This paper describes the new Scoring algorithm, very first version, to predict the cause of incident occurrences based on the various occurrence factors and impact factor.The algorithm has been trained and made learning with the help of FACTS incidents database which has the records of worldwide industrial incidents occurred between 2004 and 2014. With the help of Apache Spark, fast unified analytics engine and Microsoft .Net framework, the verification and validation of the algorithm has been done and the results of prediction have been discussed in this paper.

Workplace Incidents, Incident Occurrences; Scoring Algorithm, Predictive Model, Chemical industries incidents;Occupational Safety

"New Scoring Algorithm for predicting the cause of incident occurrences in chemical industries", International Journal of Emerging Technologies and Innovative Research (www.jetir.org), ISSN:2349-5162, Vol.6, Issue 5, page no.357-361, May-2019, Available :http://www.jetir.org/papers/JETIR1905D50.pdf

"New Scoring Algorithm for predicting the cause of incident occurrences in chemical industries", International Journal of Emerging Technologies and Innovative Research (www.jetir.org | UGC and issn Approved), ISSN:2349-5162, Vol.6, Issue 5, page no. pp357-361, May-2019, Available at : http://www.jetir.org/papers/JETIR1905D50.pdf