Data driven simulation model for the scheduling of offshore jacket construction activities.

Over the course of an Engineering, Procurement and Construction (EPC) project for the construction an offshore jacket, many uncertainties exist for both the client as well as for the contractors and sub-contractors in multiple aspects of the project. If not regulated properly these uncertainties may eventually result in waste during the construction phase.

Waste encompasses underutilization of resources, tasks not being performed on time or having increased project duration, ultimately leading to unnecessary costs.

The main objective of this thesis is to determine and address the root-causes for waste. The research incorporated a detailed analysis of the business processes regarding all phases of the EPC project and the current regulation of its tasks and requirements. Conclusion: increase in adaptability, detail and transparency is needed via a holistic approach in order to improve the regulation of processes.

Therefore, an innovative methodology is developed that consists of a framework of Building Information Models, an Enterprise Resource Planning system and a simulation model. This framework allows for the capturing of all project information, share it between involved parties automatically and use it in the simulation to (re)schedule the required activities automatically. The simulation model in particular: 1) allows integrating of all aspects of the project and their relations via model logic in a flexible way, allowing for the scheduling tasks and their requirements in a highly detailed and extensive way automatically

2) enables a transparent output of an extensive and detailed project schedule including operational data that allows for project optimization. Thereby increasing transparency.

Via a case study the project dynamics are analyzed and the operational aspects are optimized in experiment steps. These steps included testing the effects of changing space capacity, material delivery, working shifts, resources, task precedencies in the project and disturbances regarding crew-sickness and material delay. The transparent output showing these effects include changes in project lead-time, utility rates, task duration, task distribution and resource distribution.

The developed scheduling run takes 5 minutes, whereas the currently used method takes 4 weeks and is much less detailed and extensive. The methodology used as an analysis tool, revealed that the impact of changes and disturbances in a project depend on how they affect the critical path (successive string of tasks with the longest lead-time in the project). For this project the paint-shop and material arrival are the bottlenecks. Due to these bottlenecks and other limited resources, the sequence in which to perform tasks and prioritize resources is an important factor in producing the shortest critical path and therefore project lead-time. The analyses also revealed that utilization rates are increased and effect of sickness is reduced by balancing resource distribution.

This methodology can provide continuous, direct, transparent and up-to-date insight at an overarching and detailed level. This enables better understanding and monitoring. It provides the potential for manual optimization. These aspects combined improve regulation of the construction process including the reduction of waste.

It is recommended that model logic is further validated, optimization functions, uncertainties and costs are included in the model. Furthermore the integration of processes can be expanded (e.g. engineering), improving the projects’ holistic scheduling approach.

The most promising outlook is the possibility of the framework to be integrated with tablets providing and collecting data. This gives the framework real-time feedback and allows for an extended function of tracking KPI’s and updating logic in the model automatically, improving it continuously.