To　improve　the　resource　utilization　of　complex　hardware/software　systems,　a　fuzzy　task-resource　clustering　model　was　proposed　in　response　to　real-time　tasks,　using　the　average　distance　to　show　the　membership　of　fuzzy　sets.　The　fuzzy　similar　coefficient　was　defined.　In　addition,　the　Resource　Clustering　Algorithm　(RCA)　was　designed　to　cluster　real-time　tasks　and　historical　tasks,　and　according　to　clustering　results,　the　systems　would　analyze　resources　occupation　and　respond　to　real-time　tasks　rapidly.　Through　the　simulation　test　on　a　certain　complex　hardware/software　system,　the　result　shows　that　the　proposed　algorithm　can　be　used　effectively　in　the　resources　scheduling.