The　concept　of　sustainable　development,　which　strives　to　protect　the　environment　while　developing　economic　and　industrial　sectors,　has　received　widespread　attention.　This　paper　takes　the　fabric　dyeing　process　in　the　textile　industry　as　an　example　to　study　the　problem　of　minimizing　the　total　pollution　cost　and　the　maximum　lateness　on　non-identical　parallel　batch　processing　machines　with　different　unit　pollution　costs.　This　paper　analyzes　the　characteristics　of　the　batch　scheduling　problem　and　proposes　a　method　of　constructively　generating　the　initial　solutions　based　on　these　characteristics.　Meanwhile,　An　environment　selection　based　on　hierarchical　clustering　is　proposed.　This　paper　develops　an　NSGA-CC　algorithm　combining　the　above　two　methods　in　the　NSGA-II　framework　to　solve　this　bi-objective　problem.　Through　extensive　simulation　experiments,　NSGA-CC　is　compared　with　several　latest　algorithms,　and　then　the　experimental　results　show　that　the　proposed　algorithm　outperforms　compared　algorithms.