A New Optimization Technique And The Foraging Strategy Of Escherichia Coli
1663 Words7 Pages
In 2002, a new optimization technique was proposed by Passino which is inspired by the foraging strategy of Escherichia Coli (E. Coli) bacteria present in human intestines called Bacteria Foraging Optimization Algorithm (BFOA) . It is a population-based stochastic search algorithm that has been introduced to solve the problem related to optimization and control system. Since its inception, BFOA successfully has drawn the attention of many researchers from diverse fields to exploit its performance as a high-performance optimizer and has been successfully applied in real world applications such as optimal power control , image processing , jobs scheduling,  and etc. The advantages that motivate researchers to explore its…show more content… By sending the signal, it enables an individual bacterium to communicate with others. Healthy bacteria will be reproduced and poor foraging bacteria will be eliminated. The bacteria will keep repeating these processes in their lifetime.
In BFOA, each of the individual bacteria in the search space is representing an individual solution to the optimization problem . Each bacterium will undergo chemotactic steps to the direction of minimum fitness function (rich in nutrients). During the taxis, each bacterium will communicate with other to swarm in the group toward the global optimum. Bacteria will be evaluated again according to their health and sorted in ascending order. Half of them with better health will be reproduced by splitting into two and the other half of poor health bacteria will be eliminated from the search space. In order to explore more space, some of the bacteria will be eliminated and reinitialized randomly to explore unvisited space in order to find the global minimum or maximum point. For better understanding, this algorithm mechanism will be explained in solving an optimization problem.
In optimization problem that we need to find the minimum of J(θ), θ ∈ ℜp, where we do not have measurements or an analytical description of the gradient ∇J(θ). This problem is considered as a non-gradient optimization problem. BFOA does not rely on the gradient function to operate but use concentration of location of search space as the fitness function. Let θ be