Fuzzy Linear and Repetitive Scheduling for Construction Projects

Abstract

Unavoidable risks that cause uncertainty within activity durations should be modeled to gain realism in scheduling. While approaches exist for one-dimensional network schedules, two-dimensional linear and repetitive schedules that track both work and time lack such a method. Fuzzy logic can express variability by modeling optimistic, realistic, and pessimistic cases, which, in this study, form a cone of expected activity durations. Previous studies attempted to apply it to linear and repetitive schedules but suffered from inconsistencies that rendered their results incorrect. Therefore, this research proposes a new fuzzy scheduling method. Its contribution is threefold. First, it models linear or segmental activities as continuous yet flexible fuzzy cones. Second, it explicitly expresses continuous buffers in both the time and work/space dimensions to compose an efficient schedule from such inputs. Third, it identifies their full or partial fuzzy criticality and fuzzy float so that managers can split their attention accordingly. It is illustrated by a validation example, which also explains prior inconsistencies. The industry thus gains a new method that considers uncertainty to generate stable and efficient schedules.