Project management is becoming increasingly complex due to a shift of contractual responsibilities to contractors, broader project scopes with associated increase in interfaces, and the increasing influence of (local) stakeholders. This calls for adaptive decision support both for the planning and execution phase in order to find the best fitting solution for multiple (and sometimes changing) objectives. A stochastic simulation and multi-objective optimisation framework (Odycon - Open Design & DYNamic CONtrol) has been developed that enables optimisation of both strategic planning and/or dynamic control cases. Odycon is a decision-support framework that automates the selection of planning and control variables, taking into consideration multiple stakeholder objectives and constraints. To enable this, both Monte-Carlo simulation (MCS) and the multi-objective preference based IMAP optimisation are integrated. Odycon takes a next step in computer aided design for operations management into the future.
To this end, two Python based models were developed: one for a pure strategic planning application in an offshore transport and installation project (see SYPL), and another for a pure mitigation control application in an inland-infrastructure construction project (see MICO). Both applications prove their advances towards concurrent and associative design and decision-making, offering best fit-for common purpose synthesis for different complex project phases.
The optimisation is based on the Preferendus principles (Wolfert, 2023; Van Heukelum et al., 2024). For preference aggregation as part of the optimisation, the A-fine Aggregator algorithm is used.
The Odycon concept, and the associated Python models are developed by Lukas Teuber, with the help of Harold van Heukelum and Rogier Wolfert.
This repository is licensed under the MIT license.