Abstract: Computer networks are sensitive systems and are prone to error. Every time there is an error in a computer network it needs to be solved at the soonest possible time so productivity will not be affected. One problem encountered in diagnosing an error is we do not know it’s possible cause and because it is unknown, fixing the problem takes a lot of time. Trial and error is often employed to diagnose the problem. The predicament with trial and error is instead of fixing the problem it might make the problem worse. Knowing the possible cause of the problem before hand saves a lot of time in diagnostics. One tool that can be used to find the possible cause of problems in computer networks is an expert system. This system simulates human experts in diagnosing the problem. The problem with expert systems is that there may be multiple rules and the system may not know which one to fire. This research tries to solve that problem by applying the Particle Swarm Optimization (PSO) to the rules of an expert system, so it can give Impasse Weights (IW) to the rules and determine which rule is to fire. The conflict resolution algorithm for this research was tested on sample data of the problems encountered in computer networks. This research showed that particle swarm optimization can be used for an expert system conflict resolution.