The topic of this paper is the usage of the algorithm of marine predator (MPA) in the optimization of the suspension of rail vehicles with coil springs. The aim is to reduce the mass of set of coil springs as the main parts of rail vehicles suspension. The optimized set of coil springs must satisfy the appropriate conditions related to the suspension characteristics, with the aim of achieving the required operation quality and running security of the observed rail vehicle. Starting from the bi-linear characteristic of rail vehicles suspension and the analytical description of its parameters, an optimization problem is formulated. It is composed of six optimization parameters, an objective function and 16 constraints (eight for each coil spring in the set). The developed approach is applied in two specific examples of suspension optimization of four-axled freight wagons, the first with axle load of 200 kN and the second with axle load of 225 kN. The optimization problem is resolved using MPA. The acquired results showed that the given optimization approach with MPA provide a significant mass decrease compared to conventional design method of rail vehicles suspension with coil springs. The mass decrease of one set of coil springs in both examples is about 15.5%. Given that each considered four-axled freight wagon has 16 sets of coil springs, the decrease of the total coil springs mass per wagon in both examples is more than 60 kg. Since certified rail vehicles are commonly produced in large series, the proposed approach can be very significant for increasing profitability in the rail vehicles industry.