In this paper, we study a novel spatial crowdsourcing system where the workers' time availabilities and their spatial locations are known a priori. Consequently, the tasks assignment to workers is performed not only based on the current location of the human workers and the tasks available in the region, but also based on the availability of the workers during the specific times that a given task should be accepted, processed, and completed. Having the system determine the daily pattern of the workers (either by predefined questionnaires when the workers register, or by archiving data from the worker's mobile devices, or by on the road and real-time entered status data) eliminates many unsuccessful task assignments and therefore significantly increases the efficiency of the system. In the original Spatial Crowdsourcing (SC) framework, the SC-server optimizes the task assignment locally at every instance of time and whenever a new task, or a new worker, enters the system. Our new framework (PSC), on the other hand, allows the users to enter their daily routine, and temporal, spatial, and availability patterns a priori. This makes the system much more stable and pattern-opportunistic. The PSC servers can focus on receiving and archiving new entries (e.g., workers, tasks, and their criteria) during busy times (e.g., when there are many new entries in the system), and can focus on optimization and computations during quiet times (e.g., when there are fewer new entries in the system). Having the task optimization process happen during quiet times, and when there are few changes to the system, makes the performance more stable and reliable. It also allows the PSC system to have a global view of the system and and perform global optimizations to improve the performance.