This thesis considers a scheduling problem in the assembly-like flow system(ALFS) where each product is made through the sequential process of subassembly and assembly operations. The problem is investigated in two different models of minimizing the total weighted flow time, one in a simple ALFS consisting of two parallel machines and the other one in a generalized ALFS consisting of multiple parallel machines. Both the models are proved to be NP-complete, so that some dominance solution properties are respectively characterized to derive greedy-type heuristic solution procedures. The heuristic procedures are then tested for their efficiencies with various numerical examples.