Computational Intelligence Approach to Optimising Structural Configurations of Multi-Span Bridges for Rural Road Networks in Northern Nigeria

The optimal configuration of multi-span bridges for rural road networks in northern Nigeria involves a complex multi-objective design space encompassing structural efficiency, construction cost, construction material availability, maintenance accessibility, hydraulic performance, and local geotechnical constraints that resists solution by conventional analytical optimisation methods. This study develops a computational intelligence approach to structural configuration optimisation for multi-span bridges in rural northern Nigeria, combining multi-objective evolutionary algorithms with physics-based structural performance constraints and locally calibrated construction cost models. The optimisation framework integrates a non-dominated sorting genetic algorithm II for Pareto front exploration with a structural analysis engine computing member forces, deflections, and foundation reactions for parametrically defined bridge configurations. Design variables encompass span count and lengths, structural system type from a catalogue of prestressed concrete beam, composite steel-concrete beam, and reinforced concrete box culvert options, deck width, foundation type, and clearance height. Construction cost functions were derived from bill of quantities data from twenty rural bridge projects executed in Kebbi, Sokoto, Jigawa, and Kano States in the North West geopolitical zone under the Community and Social Development Agency programme. Multi-objective optimisation was conducted for four representative river crossing scenarios varying in hydraulic width, scour depth, and site accessibility class. The Pareto-optimal solution sets generated for each scenario provide bridge engineers with a family of structurally valid, cost-efficient configurations from which selection can be made according to project-specific priorities. The original contribution is the development of the first computationally grounded bridge configuration optimisation tool calibrated to rural northern Nigerian construction conditions. Keywords: computational intelligence, bridge optimisation, rural roads, Northern Nigeria, multi-objective evolutionary algorithm.