Multi-objective adaptive intelligent water drops algorithm for optimization & vehicle guidance in road graph network

In this paper we have applied Multi-Objective Intelligent Water Drops Algorithm (MO-IWDA) for optimized route determination of the vehicle through all the underutilized paths available in a road graph exploiting optimization of dynamic parameter based path planning for the vehicle users. Due to the tendency of all the vehicles to follow the same path and also the shortest distance being the implied choice, there occurs tremendous increase in waiting time and high congestion which leads to pollution and stress. Hence it becomes necessary to revise the traffic management system and with the emergence of advanced smart gadgets it is now possible to locally investigate the network information and provide the best decisions for the users so that there is not only enhancement of time factor but also there is optimal usage of fuel and power. IWDA has emerged as a successful graph based meta-heuristics and has been used in NP-hard combinatorial problems like travelling salesman problem, knapsack problems etc. The success of IWDA lies more on probabilistic exploration as there is time linked cooperation and communication between the droplets in terms of its velocity and carried sand. Here we have involved multiobjective optimization of distance and waiting time minimization and used non-weighted fitness function for decision making and evaluation of the best combinations. Here another problem has been resolved of whether there is requirement of keeping different sand quality for both distance and waiting time or the same sand can be dealt with both. A comparative study of the two schemes will clearly reveal the individuality of the IWDA in handling multi-parameter optimization or there is requirement of different types of sand types for each kind of optimization. Analysis of results shows that the optimization level depends on sand investigation criteria which are dependent on how the parameter scaling is done. For multi-objective optimization the relative ratio of scaling - s the prime factor. However due to the introduction of the exploration and adaptive parameters the algorithm will sought out the explored path and lead to global optimization.