Smart home scheduling is regarded as an effective technique in Demand Side Management(DSM). In this paper, we intended to formulate an efficient algorithm to schedule smart home appliances to minimize the monetary cost of electricity as possible. Discrete power levels for appliances and quadric pricing model are adopted in this work. The proposed algorithm consists of two parts, with cuckoo search optimization(CSO) via Levy flight searching for real number solutions as relaxed forms, and branch and bound method mapping relaxed forms to discrete power level values. Simulation results show that our algorithm outperforms the previous work in monetary cost within similar execution time.