The world of logistics and transportation can sound very arid in terms of creativity when it comes to solving problems and offering solutions. Images comes to mind of trucks full of shipping containers from different countries, entering enormous warehouses to unload thousands of products helped by a swarm of workmen circulating like little ants between high shelving units full of boxes, while some brainy engineers, locked in their dark offices and surrounded by calculators and set squares, try to bring order to so much disorder. A clear example of a complex logistics problem is e-commerce, which has caused companies to completely change their organizational strategies in order to respond almost immediately to a customer’s purchase. Nowadays, the majority of companies have an online sales channel, either direct or through an intermediary (e.g. Amazon) but the most important thing is to deliver the right product at the agreed time and at the lowest possible cost. We consumers are surprised by how these companies manage to deliver us a small purchase we have made in less than 24 hours, and even the same day, when they are receiving hundreds of thousands of orders at the same time, each one different from the other and with a completely different destination. This causes the problem to grow exponentially. Nowadays, the large retail businesses use metaheuristic techniques to solve the logistical challenges that have emerged with the new information technologies.
Traditional optimization techniques cannot provide efficient solutions to these problems, since they are usually very slow and complicated to program. Well, it is the metaheuristic techniques which allow this to happen. Indeed, today the solving of logistical problems has created a new scientific methodology, new algorithms known as Metaheuristic algorithms. These require a highly creative mind that applies both modes of conscious functions of the two hemispheres of the brain, moving between what is holistic and sequential, between intuition and logic, between the broad field of a mastery of the problem and a small clear segment specific to a particular area.
During a creative problem-solving process, it is best to begin with a divergent thought process in order to generate as many ideas or solutions as possible, later switching to a convergent thought process to select the most promising ideas. And this is precisely what we do when we develop a metaheuristic algorithm to solve a problem. Indeed, a metaheuristic algorithm is formally defined as an iterative generation process which guides a subordinate heuristic by intelligently combining different concepts for exploring and exploiting the search space, and learning strategies are used to structure information in order to efficiently find optimal or near-optimal solutions to new highly-complex problems that have arisen in the 21st century. And all of this while seeking the fastest possible execution in order to obtain results in just seconds!
There are many approaches to metaheuristic algorithms. Single solution approaches focus on modifying and improving a single candidate solution. Single solution metaheuristics include simulated annealing, iterated local search, variable neighborhood search and guided local search. Another category of metaheuristics is that based on nature, on the intelligence of swarms, which is a collective behavior of decentralized, self-organized agents in a population (or swarm). We have ant colony metaheuristic algorithms, particle swarm algorithms, social cognitive optimization algorithms, penguins search optimization algorithms and artificial bee colony algorithms, evolutionary computation and genetic algorithms, among others. Even in the name you’ve got to be creative!
