Versatile multi-objective genetic optimization for non-identical multi-storey building projects

Abstract

Construction companies face ever increasing pressure to implement projects as quickly and less expensive as possible. This paper proposes a new efficient technique to schedule several kinds of repetitive construction projects with versatile multi-objective optimization. Non-identical repeating units (floors) are commonly found. The line of balance (LOB) technique is well suited for scheduling such repetitive projects. It maintains resource continuity for an activity from one unit to the next one and achieves logic constraints at the same time. However, LOB is faced by several practical challenges when used for high-rise buildings and habitation projects: variability in construction works, diverse directions in activities workflow, skeleton constraint, and presentation difficulties. The line of balance based technique is adapted in order to overcome these challenges and maximize the performance of scheduling multi-storey building projects. The present model enables planners to control time and/or cost optimized plan using Genetic Algorithms. Activities' alternatives are investigated to generate a set of crew formations and interruptions if allowed. Sensitivity to cost attributes for cost optimizations is also handled. Spreadsheets are developed to demonstrate the model's usefulness on an example application, adapted from literature. However, other examples handled in literature are investigated to validate the proposed model.