<p>Sequence-dependent setup and deterioration are two highly prevalent phenomena in practical production processes, both of which exert a significant impact on production scheduling performance. When these two phenomena interact and couple within a production workshop, their combined effect on production performance becomes even more pronounced. However, existing research has not yet explored the coupling effect of these two phenomena in the context of production scheduling. For the first time, this paper addresses an energy-oriented job-shop scheduling problem that involves the interactive coupling effect of sequence-dependent setup and start-time-dependent deterioration. Given the multi-objective nature of this problem, a novel and efficient multi-population multi-objective whale optimization algorithm is proposed to tackle this new scheduling challenge. A hybrid population initialization strategy, integrated with problem-feature-based heuristic rules, is developed to enhance both the quality and diversity of initial solutions. Leveraging grey entropy parallel analysis and cluster analysis, a multi-population collaborative search mechanism is designed to facilitate the collaborative evolution of multiple sub-populations. Additionally, a problem-feature-based local search strategy with an adaptive learning mechanism is put forward, which dynamically selects an optimal local search operator for each sub-population based on the historical performance improvement rate. Comprehensive comparative experiments are conducted to investigate the interactive coupling effect of sequence-dependent setup and deterioration, as well as the impact of the key components of the proposed algorithm. The efficacy of the proposed algorithm is assessed via comprehensive computational experiments, with results showing that it outperforms four prominent multi-objective algorithms.</p>