Crowdsourcing platforms provide powerful means to execute queries that require some human knowledge, intelligence and experience instead of just automated machine computation, such as image recognition, data filtering and labeling. With the development of mobile devices and the rapid prevalence of smartphones that boosted mobile Internet access, location-based crowdsourcing is quickly becoming ubiquitous, enabling location-based queries assigned to and performed by humans. In sharp contrast of existing location-based crowdsourcing approaches that focus on simple queries, in this paper, we describe a crowdsourcing process model that supports queries including several crowd activities, and can be applied in a variety of location-based crowdsourcing scenarios. We also propose different strategies for managing this crowdsourcing process. Finally, we describe the architecture of our system, and present an experimental study conducted on pseudo-real dataset that evaluates the process outcomes depending on these execution strategies.