Real-time and Dynamic Planning and Scheduling Under Disruption Using Multi-hybrid Simulation and Optimisation
Files
Date
Authors
Supervisor
Item type
Degree name
Journal Title
Journal ISSN
Volume Title
Publisher
Abstract
Planning and scheduling are the principal components of construction contract execution. Distinct studies have developed a wide range of tools and technologies for solving construction planning and scheduling problems. The distinguishing feature of construction planning and scheduling is its complex and dynamic characteristics. But, most studies in construction planning and scheduling have only focused on operational and static parameters without the consideration for strategic and tactical variables. This research proposed a new dynamic planning and scheduling solutions that integrated operational, tactical and strategic parameters. The implementation followed two pathways: 1) comprehensive literature review on methods in planning and scheduling; 2) construction project time factors (CTF) influence on construction project time management (CTM) and; 3) multi-hybrid modeling, simulation and optimisation. The literature review with information gathered from the case study company (CSC) identified the need to integrate planning and scheduling with its dynamic and complex characteristics. A combination of data from the joint literature and CSC for the case study (Multi-million $ road improvement contract) depicted the need to connect the system components and characteristics. The second segment of the study investigated the influences on construction time management (CTM) using a dynamic Bayesian network (DBN). Although the DBN analysis created significant insight into CTM influences to support decision-making, the study conducted an in-depth investigation to integrate the dynamic and complex characteristics of the multi-component systems in the construction project. In the multi-hybrid model, a dynamic framework of agent-based simulation (ABS)-discrete event simulation (DES)-discrete-rate simulation (DRS)-system dynamics (SD) and scatter search optimisation was developed. The dynamic framework used the bidding phase and construction phase that formed the case study from CSC. In the stages, the mechanism of the investigation comprises operational, tactical and strategic planning. The construction phase is the construction of the road improvements contract in 33 different locations (multi-case-in-one-case). Groundwater table and access egress factors influenced the construction operations. The first idea of the research is the integration of the framework as a unit. The multi-method and dynamic framework were integrated and implemented in AnyLogic software with root in java script. After determining the framework and model integration, internal validation of the model was performed with real data collected from the CSC construction operations. The internal validation confirmed 99 % confidence between the model data and the real data. The research created variation and sensitivity experiments of the model with the influences. The scatter search optimisation experiment was executed to generate an optimal and feasible solution of the defined objective from current and infeasible solutions. A Gantt Java-class was established in each activity to gather the time spent to execute specific tasks and to carry out the scheduling solution, which was exported into Microsoft Project. External validation was implemented to get feedbacks from the CSC. The positive feedback gives the framework and its resolute solutions acceptance for practical application for predictive and reactive dynamic planning and scheduling in the construction enterprise. The feedback from CSC about the developed tool has recommended the application of the system in one of their project at the design phase.