Integrated land use and transportation models are helpful when policy, planning, or environment impacts are being evaluated, but the strengths and limitations in these models must be optimized. To optimize the ITLUP (Integrated Transportation and Land-Use Planning) model and apply it in small- and medium-sized cities in China, this study considered the constraints of land use intensity and introduced two critical indicators (the maximum number of households and maximum employment) to characterize the land capacity and improve the practicality of the model. Then, Monte Carlo simulation analysis was used to analyze the uncertainty factors using the coefficient of variation (C.V) and standardized regression coefficient (SRC). The results suggest that the maximum future employment and households may exceed the land limit and must be adjusted to a new zone, and the model operation simulation was closer to the actual situation of small- and medium-sized cities. The C.V value of the model output showed the increasing trend of the uncertainty of the model output variable over time, especially affected by DRAM model parameters, traffic demand forecasting model parameters and the peak hourly flow ratio. Such findings are meaningful for policymakers, planners, and others when the ITLUP model is used to anticipate the zonal employment and household allocation and to further explore the interaction between land use and transportation.