An Evolutionary Multi-objective Optimization algorithm for the routing of droplets in Digital Microfluidic Biochips

Many laboratory scale biochemical processes, including clinical diagnostics are being revolutionized by Digital Microfluidic Biochips (DMFBs). This is owing to their high automation capability, low cost, portability, and efficiency. Central to the efficient operation of these devices is the droplet routing problem that aims to drive a set of droplets, each from its source to its target cells, without violating a given set of fluidic and timing constraints. The efficiency of the routing is measured by the amount of cells used and the arrival time of the latest droplet and both criteria are aimed to be minimized simultaneously. To solve this problem we propose an Evolutionary Multi-objective Optimization algorithm for the Droplet Routing problem (EMO-DR) based on the NSGA-II framework, where the crossover operator is not used. EMO-DR features new mutation operators and a biased random generator of initial solutions. Experimental results show that the proposed approach produces competitive results when compared with those obtained through state-of-the-art methods. The paper also highlights the main challenges that evolutionary approaches need to solve when dealing with this routing problem.