Mushroom Reproduction Optimization (MRO): A Novel Nature-Inspired Evolutionary Algorithm Mahdi Bidar Department of Computer Science University of Regina Regina, Canada bidar2m@uregina.ca Hamidreza Rashidy Kanan Department of Computer Engineering Shahid Rajaee Teacher Training University Tehran, Iran h.rashidykanan@srttu.edu Malek Mouhoub and Samira Sadaoui Department of Computer Science University of Regina Regina, Canada {sadaouis,mouhoubm}@uregina.ca Abstract—We introduce a new nature-inspired optimization algorithm namely Mushroom Reproduction Optimization (MRO) inspired and motivated by the reproduction and growth mech- anisms of mushrooms in nature. MRO follows the process of discovering rich areas (containing good living conditions) by spores to grow and develop their own colonies. We thoroughly assess MRO performance based on numerous unimodal and multimodal benchmark functions as well as engineering problem instances. Moreover, to further investigate on the performance of the proposed MRO algorithm, we conduct a useful statistical evaluation and comparison with well known meta-heuristic algo- rithms. The experimental results confirm the high performance of MRO in dealing with complex optimization problems by discovering solutions with better quality. I. Introduction Optimization is the process of finding a scenario that minimizes or maximizes an objective function while satisfy- ing a given set of constraints. This scenario, called optimal solution, is often searched in an exponential set of candidate solutions, and therefore requires a very expensive execution time. To overcome this difficulty in practice, meta-heuristic approximation methods have been introduced. While these solving methods do not guarantee the solution optimality, they are however very successful in returning near-to-optimal solutions [1], [2], [3]. Two strategies are used by meta- heuristic algorithms to find the best solution, exploitation and exploration synonymous with intensification and diversifica- tion respectively. Generally speaking, exploitation looks for the best solution within the local scale whereas exploration searches globally for the solution in the problem space [4]. Exploration prevents meta-heuristic algorithms from being stuck in local optima. These algorithms have been widely applied to high-dimensional optimization problems, such as optimization of objective functions [5], operation and control of power systems [6], job scheduling and assignment [7], [8], [9], pattern recognition [10] and image processing [11]. In this paper, we focus on nature-inspired meta-heuristic algorithms that can be divided into three categories according to their source of inspiration [12]: a) Swarm intelligence algorithms, b) Non-swarm intelligence algorithms, c) Physical and chemical algorithms. The swarm intelligence class refers to the collaborative behaviour of agents (often unintelligent) of decentralized and self-organizing swarm systems. By following simple govern- ing rules to perform simple tasks for reaching the goals, the collective intelligence is attained. The most well-known algorithms belonging to this category are: Ant Colony Opti- mization (ACO), inspired by the foraging behaviour of real ants, Particle Swarm Optimization (PSO), inspired by the swarm intelligence of birds and fishes, Honey Bee Colony Op- timization (HBCO), based on honey-bee behaviour in finding food sources, Firefly Algorithm (FA) which mimics the social behaviour of fireflies and their light emitting characteristics, Bat Algorithm (BA) inspired by the echolocation behaviour of bats, Cuckoo Optimization Algorithm (COA) inspired by the cuckoo characteristics in laying eggs and breeding, Krill Herd (KH), based on the herding behaviour of krill individuals in their environment, Grey Wolf Optimizer (GWO), inspired by the leadership hierarchy and hunting mechanisms of grey wolves, and Ant Lion Optimizer (ALO) which mimics the hunting mechanism of real ant lions. On the other hand, non-swarm intelligence algorithms are not endowed with the collaborative behaviour of their agents. The popular algorithms of this category include: Genetic Algorithms (GAs) based on the Darwinian evolution and theory of survival of the fittest, Differential Evolution (DE) which consists of improving a potential solution w. r. t. a given measure of the solution quality, Invasive Weed Optimization (IWO), simulating the ecological behaviour of weeds in finding places for growth and reproduction, and Biogeography-Based Optimization (BBO) which borrows ideas from island biogeography. The third category of algorithms is inspired by the physical or chemical processes, and the most famous ones are: Gravitational Search Algorithms (GSAs), based on the law of gravity and mass interactions, Interior Search Algorithms (ISAs), inspired by the interior design and decoration, and Central Force Optimization (CFO), inspired by metaphor of gravitational kinematics. Al- though the aforementioned nature-inspired algorithms provide satisfactory results, however an algorithm may be successful in certain optimization problems but not in others [13]. Thus, developing a new optimization method to deal effectively with a wider range of problems as well as to tackle efficiently more complex problems is still a recognized research objective [13] [14]. Our aim in this present study is to elaborate a novel optimization algorithm inspired by the real mushroom