Enhanced Constraints Representation and MTSP Schedule Optimization for Repetitive Projects Considering Environmental Sustainability

Abstract

The Canadian construction industry is facing serious challenges regarding project delays (affecting 25% - 50% of large infrastructure projects), cost overruns (totaling around $91 billion between 2017 and 2023), and contributing nearly 39% to Canada’s annual Greenhouse Gas emissions. Most of the infrastructure projects, e.g. bridges, highways, and multi-school rehabilitation, involve tasks that are repetitive in nature. Thus, repetitive scheduling techniques offer multiple benefits, including economy of scale and learning curve savings. However, planning such projects remains complex, particularly when units are not identical in size and multiple crews are being involved to fast-track the project. In response, this research developed novel Multi-Travelling Salesman Problem (MTSP)-based schedule optimization model(s) offering an enhanced constraints representation for optimizing key repetitive construction projects. The proposed research framework incorporates three modules to handle various types of repetitive projects (linear and scattered). It also provides more sustainable schedules by integrating environmental impacts into the optimization process. Unlike existing models, the proposed modules formulate the scheduling problem as a MTSP where crews move across units to complete their assigned tasks while meeting time and budgetary constraints. One of the modules also quantifies, monetizes and incorporates key environmental impacts into the optimization process, accounting for variations in construction methods and materials and their effects on time, cost, and sustainability. The optimized schedules are then visualized through legible charts, GIS maps, and intuitive construction simulations. This modules formulation is unique. Its novelty beyond existing repetitive scheduling models stems from: (1) handling multiple crews simultaneously by incorporating crew routing sequences as key scheduling variables through a MTSP formulation; (2) introducing adjacency constraints to maintain, for all crews, both work continuity and unit linearity, if needed, while accounting for transition times between units; (3) adding new criteria to the multi-mode construction scheduling models where each task can be executed using optimal alternative methods and materials affecting time, cost and environmental considerations; and (4) generating sustainable construction schedules by quantifying and integrating environmental impacts into the optimization process. The proposed model is validated against previous models in the literature through several case studies. Not only do these modules provide more efficient computational performance, but they also demonstrate significant improvements across various optimization objectives, such as project duration, crew transition-time, crew idle-time, environmental impacts, total cost, and various combinations. The proposed research framework provides significant benefits as a scalable and adaptable solution for managing complex repetitive projects, enhancing the delivery of infrastructure projects and offering substantial benefits to the multi-billion-dollar construction industry.