Mahajan Neha et al.; International Journal of Advance Research, Ideas and Innovations in Technology © 2019, www.IJARIIT.com All Rights Reserved Page |232 ISSN: 2454-132X Impact factor: 4.295 (Volume 5, Issue 3) Available online at: www.ijariit.com Prediction of building structure age using machine learning Neha Mahajan mahajanneha999@gmail.com Usha Mittal Institute of Technology, Mumbai, Maharashtra Darshana Patil darshpatil12321@gmail.com Usha Mittal Institute of Technology, Mumbai, Maharashtra Apurva Kotkar apurvakotkar96@gmail.com Usha Mittal Institute of Technology, Mumbai, Maharashtra Kumud Wasnik kumudwasnik@gmail.com Usha Mittal Institute of Technology, Mumbai, Maharashtra ABSTRACT Determining the service life of building structure is a critical step for the evaluation of maintenance of the building. If the service life of a building is known then proper maintenance and refurbishment steps can be taken. Previous studies focused only on physical obsolescence whereas new concept focused on other new six criteria. The objective of this paper is to use the entire six criteria for the evaluation of obsolescence for the prediction of the age of building a structure using machine learning. A prediction model for predicting age is developed by combining the six obsolescence criteria, absolute weights, diagnostic scores, and machine learning. As obsolescence is predicted using all six criteria, the manual calculation is reduced to provide more accuracy. This prediction model has built using python language, Django as a framework and PyCharm IDE. To make this prediction system more adaptable the website is created and hosted on webhost000.com by combining prediction model and UI of the website which displays the predicted age for respective diagnostic scores. Hence predicting age is useful for taking actions to prevent future degradation. Keywords— Service life, Obsolescence, Diagnostic scores, Prediction model, PyCharm IDE, Django, Adaptable 1. INTRODUCTION Nowadays, we have heard many cases of infrastructure collapsing. The recent example is the Andheri Bridge. This is because no proper attention is paid towards the maintenance of infrastructure. Although building are long lasting, continuous maintenance and refurbishment is required. So prediction of age of building is must to predict the future chances of collapsing and thus preventing the damage. Therefore our proposed system predicts the age of the building which helps in preventing the collapsing of building. The aim of this paper is to develop a prediction model for predicting buildings age based on rate of obsolescence. To achieve the aim, implementation is divided into three modules. First, a literature review was done to decide the method of determining the age. The first module comprise of interface which is used to collect diagnostic scores. Second module has been implemented for obsolescence prediction using machine learning algorithm of Multiple Linear regression. Third module uses obsolescence value to predict age of building. Our objectives of the study are listed as follows:  To predict the lifespan by estimating the age of the infrastructure  To take preventive measures for damage  To reduce cost spend after damage  Reduce the risk factor of life 2. EXISTING SYSTEM 2.1 Using a combination of various structural materials and environments It attempts to assign useful life values for buildings having various combinations of main (structural) materials and environments. The base case is taken as reinforced concrete because that will probably be the most common building material today. The predominance of structural steel and timber is likely to be seen only in older buildings. A building will have different materials of different qualities in a variety of microenvironments. Poor quality of construction could reduce useful life by up to 20 years; this would depend on the combination of material and environment. Table 1: Main Structural Elements [6] This method does not give expected accuracy as it has considered only two parameters that is. Materials and Environment. So a system that takes into consideration various factors is required.