An Algorithm for Construction Project Cost Forecast Based on Particle Swarm Optimization-Guided BP Neural Network

Algorithm 1

Flow of particle swarm algorithm.

 Steps of specific operation

 Step 1: set parameters such as , , and , termination conditions in the algorithm.

 Step 2: initialize the population, including random positions and speeds.

 Step 3: evaluate the fitness value of the particle fitness.

 Step 4: the fitness value of each particle is compared with the best position (individual extreme value) it has passed. If the current fitness value is better, its position is taken as the current best position .

 Step 5: similarly, the fitness value of each particle is compared with the global best position it has passed, and if the current fitness value is better, its location is taken as the current global best position .

 Step 6: update the speed and position of particles.

 Step 7: determine whether the termination conditions are met, if not, return to the third step to continue the iterative update. Otherwise, the corresponding to the current fitness value is output as the global optimal solution, and the search is stopped.