In control engineering, there may be two systems that have the same input-output characteristic with different degrees of complexity. This concept leads to order diminution(OD) of a large scale system. In this article, the authors propose a new hybrid order diminution technique based on the time moment matching method with the Routh array concept and a recently developed fast and accurate salp swarm optimization (SSO) technique. The proposed method combines the advantages of both the classical method of OD and the optimization technique. The unknown coefficient of the divisor of the reduced system is obtained by exploring the time moment matching methodology with the Routh array concept, whereas the unknown coefficients of dividend polynomial are obtained by the SSO technique. The time moment matching with the Routh array ensures the nature of the system in terms of stability, and better search capability of the SSO technique reduces the error between the original and the diminished system. The proposed technique is tested on different benchmark problems, including a time-delay system. An intensive comparative study in terms of different errors, time, and frequency domain provides better performance of the proposed method compared to the existing techniques. A good match to the parameters of transient specifications indicates the success of this proposed technique. Comparatively, the matching of rise time, settling time, and maximum overshoots are 99.6076%, 99.1611%, and 100%, respectively.