Recursive heuristic scheduling method for multi-crop plant factory with solar panel roof

Abstract

Plant factory can provide stable crop cultivation environment, shorten cultivation duration, and deliver better produce quality by controlling temperature, humidity, lighting, nutrient supply, and other cultivating factors. Due to the high cost of operation, how to select crops and plan the cultivation schedule to enhance profitability is an important issue. In this research, the plant factory scheduling problem was formulated as a mixed integer linear programming (MILP) problem. The objective function is to seek the maximum revenue by considering several practical operating conditions such as (1) crop market value per unit and time, (2) crop adjacency issue, (3) cleaning and maintenance, and (4) sunlight blocking by solar panel on the roof. Although the scheduling problem can be solved by the optimization solver, it presents challenges of rack-level modification when applying the optimization method on the field. Therefore, this study further developed a heuristic algorithm, named Heuristic Plant Factory Scheduler (HPFS), which applies a recursive technique to schedule crops rack by rack. This approach provides ease of implementation and modification. The experimental results show that HPFS is able to accelerate the computational performance with acceptable scheduling quality when the problem space is large.