Spatial forest resource planning using a cultural algorithm with problem-specific information

• We propose and implement a cultural algorithm for spatial forest resource planning.
• We consider problem-specific information to design cultural/evolutionary operators.
• Our approach outperforms the previous best simulated annealing (SA) by 0.44%∼1.13%.
• Fitness ranges show that our approach obtains more stable solutions than the SA.
• SA was shown to be better than evolutionary algorithm (EA), but our new EA wins SA.

A cultural algorithm (CA) is proposed for the spatial forest resource planning problem that aims at maximizing the total timber volume harvested over a harvest planning schedule, subject to constraints of minimum harvest age, minimum adjacency green-up age, and approximately even volume flow for each period of the schedule. To increase the solution-search ability, the CA extracts problem-specific information during the evolutionary solution search to update the belief space of each generation, which has cultural influences and guidance on the next generation. The key design of the proposed CA is to propose the cultural and evolutionary operators specifically for the problem. This work is of high value as a comprehensive experimental analysis shows that the proposed CA rooted from evolutionary algorithm (EA) obtains 0.44%–1.13% better fitness and performs more stably than the previous best-known simulated annealing (SA) approach, which was shown to perform better than the EA previously.