Vulnerability assessment for disaster studies pertaining to natural hazards has evolved as a discipline in itself. The multidimensional approach for the design of a vulnerability framework is widely accepted and used, where the prevalently used dimensions are economic, social, physical, environmental. Although the dimensions of vulnerability are distinct, these are commonly aggregated together to compute the overall vulnerability of a place (i.e. composite vulnerability). It is observed this practice leads to loss of information and averaging out of scores obtained from individual dimensions of vulnerability. This study proposes a new framework for assessing multidimensional vulnerability of a region in a multi-objective framework. Individual dimension of vulnerability is computed using a new aggregator function proposed in the study. The proposed methodology has been demonstrated using the case study of a coastal district (South 24 Parganas) in West Bengal, India. It is one of the most impoverished districts of the state and has been exposed to multiple incidences of catastrophic events like tropical cyclones, storm surges and flooding. The vulnerability indices for each dimension of vulnerability were calculated using the proposed aggregator function. Pareto optimality conditions were used to obtain a Pareto frontier from where the Blocks having highest overall vulnerability were selected. This method was repeatedly used to sort the vulnerabilities of all the constituent Blocks of the district in varying levels of vulnerabilities. It was observed that Gosaba, Patharprotima, Kultali, Canning-II, Namkhana and Sagar were the most vulnerable Blocks in the district. This methodology posits that the hierarchy-based clustering system obtained from non-weighted Pareto optimality conditions is better in terms of evaluating the vulnerability of a region. It provides a system of evaluation which is free from decision maker's prejudices and guards it against loss of information - thus eliminating two weaknesses of conventional aggregation methods. It also makes it easier to arrive at specific disaster management policies.