This study proposes an Intelligent Transport System (ITS)-based smart parking or car parking space prediction (CPSP) system using Time-aware Long Short-Term Memory and Kronecker product Gated Recurrent Unit (TLSTM-KGRU) for smart cities. The method first predicts parking occupancy using Time-aware LSTM on the Kuala Lumpur Convention Centre sensor dataset, followed by Linear Interpolations and Kronecker product GRU for forecasting. Compared to SGRU-LSTM and CPSP with DELM, TLSTM-KGRU improves sensitivity, specificity, and reduces prediction time, enabling more efficient urban parking management.