The　application　of　workflow　in　office　OA　has　gradually　matured,　but　it　needs　to　be　more　real-time　and　random　in　the　field　of　industrial　Internet　of　things.　The　traditional　modeling　methods　and　process　design　can　not　meet　the　industrial　process　scenarios　in　industrial　Internet　of　things.　Therefore,　this　paper　takes　the　application　of　workflow　in　the　industrial　Internet　of　things　as　the　background,　and　aims　to　realize　the　workflow　engine　platform　of　probabilistic　timed　automata　based　on　the　industrial　Internet　of　things.　In　this　paper,　through　the　requirements　analysis,　based　on　probabilistic　timed　automata　to　establish　the　process　model　based　on　industrial　Internet　of　things,　establish　workflow　engine,　design　the　overall　idea　of　workflow　system　platform,　we　adopt　multi　module　structure　and　Spring　Boot　+　MyBatis　+　Apache　Shiro　+　Redis　+　Flowable　+　Beetl　and　other　mainstream　technologies　to　form　a　complete　set　of　workflow　engine　platform.　It　can　meet　the　requirements　of　industrial　process　and　make　the　industrial　process　more　efficient.　©　2021　IEEE.