The artificial bee colony (ABC) algorithm has such characteristics as few control parameters, strong local optimization ability and fast convergence speed. However, when solving path optimization problems, it can easily fall into local optimal solutions. In order to solve the problem of pipeline routing in a ship pipeline system, an improved artificial bee colony (IABC) algorithm is proposed.

Based on the traditional artificial bee colony algorithm, the crossover operation of genetic operators is introduced into the update mechanism of following bees, and an adaptive strategy is adopted for the crossover probability of the crossover operator. The crossover operation on the population is used to find new solutions in the global range. The way scout bees search for new paths is improved from updating the points that the path passes to updating the "road sections" in the path. This paper proposes an artificial bee colony co-evolution algorithm for solving the optimization of branch pipeline paths.

Compared with the standard artificial bee colony algorithm, the improved algorithm can improve the path layout effect by 32.3%–37.4% and the convergence speed by 17.7%–29.9%.

The improved artificial bee colony algorithm proposed herein has higher solution quality, faster convergence speed and better stability than the traditional artificial bee colony algorithm for a single pipe or branch pipe.