Aiming　at　Flexible　Job-shop　Scheduling　Problem　(FJSP)　with　the　objective　of　minimizing　the　maximum　completion　time,　the　non-uniqueness　of　process　coding　affects　the　efficiency　of　optimization　computation　through　the　recurrence　relation　of　job　completion　time.　Consequently,　a　new　model　and　the　corresponding　cross　entropy　algorithm　were　proposed.　The　unified　representation　of　solutions　based　on　Gantt　chart　was　established　to　improve　the　convergence　speed　of　probability　distribution　matrix.　Random　distribution　screen　was　introduced　to　ensure　the　feasibility　of　sampling.　By　means　of　stochastic　switching,　two　stage　method　and　cross　entropy　method　were　combined　to　speed　up　the　solution　process,　and　at　the　same　time　ensure　the　universality　of　solution　distribution.　Numerical　results　of　typical　examples　showed　that　the　new　model　had　a　significant　effect　on　the　convergence　speed,　and　the　hybrid　algorithm　could　solve　FJSP　efficiently.　©　2021,　Editorial　Department　of　CIMS.　All　right　reserved.