Swarm-intelligence collaboration based regular scheduling and dynamic rescheduling of precast component production: in prefabricated building project management

Researchers have represented a swarm-intelligence (SI) collaboration, mapping the entities and processes involved in order integration, scheduling decisions, and anomaly handling in the PC factory to eight agents, and then constructs their respective definition and collaboration modes. The dynamic-interval synergy auction (DISA) strategy and weighted Tchebycheff approach (TCH) concept are used, under multi-objective and multi-constraint conditions, to generate optimal schemes respectively. Published in Smart Construction, this approach provides the optimal conventional scheduling scheme and dynamic rescheduling scheme for PC component factories.

Prefabricated concrete structure is the fastest developing building form at present due to low construction cost, low environmental pollution, and convenient operation, so the output of precast concrete (PC) components is also showing obvious growth trend. The number of PC components with different types and complex production processes in some large factories is enormous. Some research in the Architecture, Engineering, and Construction (AEC) field indicates that intelligent approaches can enhance management efficiency and reduce costs, so the production scheduling management is necessary for PC factories.

However, due to the backwardness of management methods, most PC factories are still using inefficient and simple scheduling methods for order arrangement and production management of assembly lines, whose production often fall into dilemma when the number of orders or component types become large, or some abnormal events occur. The traditional scheduling methods mainly have the following problems:

1. The time-driven mechanism for troubleshooting at fixed intervals has a certain hysteresis, and the event-driven mechanism for response strategies maybe cause insufficient stability.

2. The traditional CNP mechanism has the shortcomings of vicious competition and local optimization, which cannot completely solve the production scheduling problems in large PC factories.

3. With the increase of the number of objectives in production scheduling, the contradiction between the diversity and convergence of the solution set intensifies. Therefore, the optimization methods that only apply to two or three objectives are no longer appropriate.

Addressing these main bottleneck issues, Sihao Li et al. from Southeast University mapped the entity objects and scheduling processes of the production to a SI collaboration network with adaptive and communication cooperation capabilities, and the mutual cooperation mode between each module is designed.

First, the dynamic-interval synergy auction (DISA) strategy is proposed to improve CNP by setting up a dynamic time window.

Second, the Tchebycheff decomposition strategy is introduced into the genetic algorithmthe to improve convergence speed and reduce computational complexity.

Third, the SI model is used to solve the optimal production scheduling scheme under different conditions such as interference-free, order change, and equipment fault respectively.

Finally, the case simulation was calculated using MATLAB 2021b to obtain feasible scheduling schemes and rescheduling schemes of the dynamic production process.

The decentralized negotiation mode with dynamic time window mechanism can avoid local optimization of schemes. Compared with traditional calculation method, this method could obtain more comprehensive and lower cost schemes. Based on the collaboration mechanism, with improved CNP and TCH strategies introduced, the dynamic model can improve the integrity and intelligence of PC factory.

This paper ”Swarm-intelligence collaboration based regular scheduling and dynamic rescheduling of precast component production: in prefabricated building project management” was published in Smart Construction.

Li S, Peng C, Chen G, Liang Y, Xu Z. Swarm-intelligence collaboration based regular scheduling and dynamic rescheduling of precast component production: in prefabricated building project management. Smart Constr. 2025(1):0002, https://doi.org/10.55092/sc20250002.