The increase in energy consumption, environmental pollution issues, and low-carbon agenda has grown the research area of demand side management (DSM). DSM programs provide feasible solutions and significantly enhance the efficiency and sustainability of electrical distribution systems. This paper classifies and discusses the broad definition of DSM based on the comprehensive literature study considering demand response and energy efficiency. The implementation of Artificial Intelligence algorithms in DSM applications has been employed in many studies to help researchers make optimal decisions and achieve predictions by analyzing the massive amount of historical data. Owing to its simplicity and consistent performance in fast convergence time, Particle Swarm Optimization (PSO) is widely used as a part of the swarm AI algorithm and has become a prominent technique in the optimization process to exploit the full benefit of the demand-side program. The variants of PSO have been developed to overcome the limitations of the original PSO and solve the high complexity and uncertainty in the DSM optimization process. The proposed PSO-based algorithm can optimize consumers' consumption curves, reducing the peak demand and hence minimizing the electricity cost when integrated with the DR programs or EE measures. The research works of the PSO algorithm in DSM have seen an increasing trend in the past decade. Therefore, this paper reviewed the application of the PSO-based algorithm in DSM fields with some constraints and discussed the challenges from the previous work. The potential for new opportunities is identified so that PSO methods can be developed for future research.INDEX TERMS Demand side management (DSM), demand response (DR), energy efficiency (EE), metaheuristic algorithms, particle swarm optimization (PSO), swarm intelligence.

In mitigating the peak demand, the energy authority in Malaysia has introduced the enhanced time of use (EToU). However, the number of participants joining the programs is less than expected. Due to that reason, this study investigated the investment benefit in terms of electricity cost reduction when consumers subscribe to the EToU tariff scheme. The significant consumers from industrial tariff types have been focused on where the load profiles were collected from the incoming providers' power stations. Meanwhile, ant colony optimization (ACO) and particle swarm optimization (PSO) are applied to optimize the load profiles reflecting EToU tariff prices. The proposed method had shown a reduction in electricity cost, and the most significant performance has been recorded congruently. For a maximum 30% load adjustment using ACO optimization, the electricity costs have been decreased by 10% (D type of tariff), 16% (E1 type of tariff), 9% (E2 kind of tariff), and 1.13% (E3 type of tariff) when compared to the existing conventional tariff. The cost-benefit of the EToU tariff switching has been identified where the simple payback period (SPP) is below one year for all the industrial types of consumers.