Oil and gas construction projects are complex and risky because of their dynamic environment. Furthermore, rising global energy demand has increased the need for trustworthy risk assessment models for such projects that can provide adequate and precise policy planning. Traditional risk assessments in oil and gas construction projects do not consider the interrelationships of factors in the best-fit models. The Decision Making Trial and Evaluation Laboratory (DEMATEL) and Analytic Network Processes (ANP), called the DEMATEL-ANP approach, have been applied to other research disciplines to address this shortcoming. This method is able to construct a structural relationship among the different influence factors to visualize complex correlations. Thus, the purpose of this study is to showcase the DEMTAL-ANP risk assessment model to assess the overall risk factors of OGC projects. This study thus identifies the crucial risk criteria of such projects. Data were collected in 2016 through interviews with experts active in OGC projects in Iran. DEMATEL in this situation is used to determine the interdependencies’ relative strengths among the risks. The ANP method is applied to assess the relative importance of the risk factors and to determine the best strategy for implementation of a risk management program. The results presented in this study are a novel adaptation of the risk assessment methodology to OGC projects that determines the important risk factors that directly affect the project success, which in turn helps in formulation of policies for ensuring reliable energy supply planning.

As a prevalent disaster, landslides cause severe loss of property and human life worldwide. The specific objective of this study is to evaluate the capability of artificial neural network (ANN) synthesized with artificial bee colony (ABC) and particle swarm optimization (PSO) evolutionary algorithms, in order to draw the landslide susceptibility map (LSM) at Golestan province, Iran. The required spatial database was created from 12 landslide conditioning factors. The area under curve (AUC) criterion was used to assess the integrity of employed predictive approaches. In this regard, the calculated AUCs of 90.10%, 85.70%, 80.30% and 76.60%, respectively, for SI, PSO-ANN, ABC-ANN and ANN showed that all models have enough accuracy for simulating the LSM, although SI presents the best performance. The landslide vulnerability map obtained by PSO-ANN model is more accurate than other intelligent techniques. In addition, training the ANN with ABC and PSO optimization algorithms conduced to enhancing the reliability of this model. Note that, a total of 76.72%, 23.96%, 30.55% and 5.37% of the study area were labeled as perilous (High and Very high susceptibility classes), respectively by SI, PSO-ANN, ABC-ANN and ANN results.

Strong ground motions usually trigger lots of slope failures in the affected area. In this work, we analyse the occurrence likelihood of earthquake-triggered landslide by employing the ensembles of adaptive neuro-fuzzy inference systems (ANFIS) with four well-known metaheuristics techniques, namely particle swarm optimization (PSO), genetic algorithm (GA), ant colony optimization (ACO), and differential evolution (DE) algorithms. Twelve landslide conditioning factors namely, elevation, slope degree, lithology, peak ground acceleration (PGA), stream power index (SPI), topographic wetness index (TWI), distance to road, distance to river, distance to fault, normalized difference vegetation index (NDVI), slope aspect, and plan curvature are considered within the geographic information system (GIS) to produce the required spatial database. In this paper, frequency ratio (FR) model is used to evaluate the spatial interaction between the landslides and conditioning factors. Meantime, among a total of 458 marked earthquake-induced landslides, 366 (80%) are specified to the learning process, and the remaining 92 (20%) landslides are used to evaluate the accuracy of applied models. The landslide susceptibility maps are generated in the GIS environment. Three accuracy criteria of mean square error (MSE), root mean square error (RMSE), and area under the receiving operating characteristic curve (AUROC) are used to develop a ranking system for comparing the integrity of the designed models. The total ranking scores (TRSs) of 15, 8, 10, and 18, respectively, obtained for PSO-ANFIS, GA-ANFIS, ACO-ANFIS, and DE-ANFIS revealed the superiority of the DE algorithm compared to other metaheuristics techniques. Also, the DE-ANFIS emerged as the fastest ensemble, due to the highest convergence speed obtained for this model.