Abhijit Baidya

Models for solid transportation problems in logistics using particle swarm optimisation algorithm and genetic algorithm

Transportation policy seeks to improve agency freight and cargo management and enhance sustainable, efficient and effective transportation operations. In this paper, four new fuzzy fixed charge solid transportation problems (FFCSTP) are formulated to maximise the total profit and minimise the total cost. The interval objective function is approximated to an intervalvalued function, i.e., transformed to a single objective using weighted sum method and weighted multiplication method. The fuzzy constraints are converted to its equivalent deterministic form using different interval order relations. Genetic algorithm (GA) and particle swarm optimisation (PSO) algorithm are used to obtain the optimal transportation schedule for the proposed solid transportation problem. During the evaluation of the models, in one case, limitation on the transported amounts is imposed and in other case, no such limitation is used. The models are illustrated with numerical examples and the optimum results of the models are compared.

Breakable Solid Transportation Problem with Hybrid and Fuzzy SafetyFactors using LINGO and Genetic Algorithm

nvironments. If we carrying the produce from sources to destination by the means of unlike conveyances then due to insurgency, land slide and bad road, there are some risks or difficulties to transport the items. By this motive we initiate “Safety Factors” in transportation problem. Due to this reason desired total safety factor is being introduced. Also our objective is to evaluate the solution of STP using expected value model. Here we develop six models where first three models are formulated taking crisp unit transportation cost but the remaining three models are formulated taking hybrid unit transportation cost. To build up the different models we consider breakability and safety factor which is taken as crisp, fuzzy and hybrid for assorted models. All the fuzzy and hybrid models are reduced into its crisp equivalent using expected value modeling. Finally by Generalized Reduced Gradient (GRG) method using LINGO.13 optimization software and Genetic Algorithm we solve the mathematical models and put a enlarge discussion on it.