An improved MILP model for double-row layout problem: Incorporating robustness, unequal areas, fixed-position machines, and walking worker operators

The double-row layout problem (DRLP) involves machines arranged on both sides of a straight-line corridor and arises in various industries. This study aims to improve the mixed-integer linear programming (MILP) model for DRLP in the literature, thereby increasing its applicability to real systems and the industry. We introduce Robust Mixed-Integer Linear Programming (RMILP), which incorporates real-world system factors such as unequal area, fixed-position machines, aisle width, material flow changes, product quantity fluctuations, and human factors, including fatigue in walking worker operators. The improvements include enhancing the objective function, variables, and certain constraints. In addition, new constraints are introduced to determine the positions of both movable and fixed facilities, along with a formulation for computing the y-coordinates. Computational simulation on real cases demonstrates that RMLIP consistently produces a single optimal layout. It achieves the lowest Total Movement Cost (TMC), with a 6.59 reduction compared to the existing layout without uncertainty and a 22.42 reduction compared to the existing layout with three products (3P) and four material flow (4MF). Furthermore, a sensitivity analysis was conducted to evaluate the impact of variations in demand and material flow on the TMC. Human factors, including operator fatigue, recovery, and movement frequency, were also examined for their effect on the total walking worker operator cost (TWOC). © The Author(s), under exclusive licence to Springer-Verlag France SAS, part of Springer Nature 2025.