A series of studies show the applicability of the NK model in the crowdsourcing research, but it also exposes a problem that the application of the NK model is not tightly integrated with crowdsourcing process, which leads to lack of a basic crowdsourcing simulation model. Accordingly, by introducing interaction relationship among task decisions to define three tasks of different structure: local task, small-world task and random task, and introducing bounded rationality and its two dimensions are taken into account: bounded rationality level that used to distinguish industry types and bounded rationality bias that used to differentiate professional users and ordinary users, an agent-based model that simulates the problem-solving process of tournament-based crowdsourcing is constructed by combining the NK fitness landscapes and the crowdsourcing framework of “Task-Crowd-Process-Evaluation”.