Delay hierarchy propagation model
Construction projects are always exposed to delay. Research has shown that most projects encounter delays and this problem is a global one. Previous research related to delays in construction projects have been dedicated to measuring and ranking the direct delays that have occurred. These types of delay are past delays and have already affected many aspects of the project's performance. This type of research is of the reactive type and handles delays after they have happened. The objective of this research is to model the construction project delays that can be used to predict the level of delays that the project could face during its future life. The proposed Delay Hierarchy Propagation Model (DHPM) is the first attempt to model delays in the construction project. This model is an innovative predictive approach to anticipate the future encountered delays before they become real. The model assumes that the direct delay is generated from earlier events or aspects that are found before the direct delay occurs; these events are called the root delay causes. These root delay causes need to be analysed, measured and managed in order to prevent or mitigate the effect of a later direct delay in the project life. The direct delays were analysed by a cause-effect technique to extract a set of root delay causes. The model assumes that the root delay causes will influence the project resources supply rate. The resource shortage then leads to activity delay and, hence possible delay to the whole project. The DHPM consists of two interrelated models: a Resource Shortage Possibility (RSP) model and the Predicting Project Delay model (PPD) model. The RSP model objective is to predict the possibility of resource shortage, whilst the PPD model objectives are to predict the project finish time and to define the critical areas for the project to delay using the output of the RSP model as input. The RSP model was verified through interview questionnaires with a number of selected personnel from the construction industry. The Delphi method was used to enhance the questionnaire results. The RSP model calculations used a combination of fuzzy logic, analytical hierarchy process (AHP) and multi-attribute theory to obtain the model output. A prototype computer program was introduced. The prototype computer program was then tested on a real construction project. The application of the RSP model showed that it is viable. The PPD model used probabilistic networking to predict the finish time of the project. The model introduced two new terms that can be used to define the most critical activities and the possible resource influence to delay. The comparison between PPD and the classical critical path method (CPM), programme evaluation and review technique (PERT) and Monte Carlo simulation revealed that the proposed model provides new information required to enhance delay management by project management staff.