This study uses ANN-based lower upper bound estimation (LUBE) method for construction of prediction intervals (PIs) at different confidence levels (CL) for the compressive strength of concrete for the first time. For the purpose of this study, an experimental study is done to prepare the required database from different mix designs of concrete. The results of this study demonstrate efficiency of the LUBE method for the three CLs considered of 85%, 90% and 95% in which the values of prediction interval coverage probability (PICP) are all greater or equal than CLs, which indicates that the ANN-based LUBE method can produce PIs with a reliable coverage probability. In addition, with average interval width index of 34.0% and the average failure distance index of 3.7% for three confidence levels, LUBE represents a more reliable and informative than exact point predictions for the compressive strength data in the test data set.

The use of data driven models has been shown to be useful for simulating complex engineering processes, when the only information available consists of the data of the process. In this study, four data-driven models, namely multiple linear regression, artificial neural network, adaptive neural fuzzy inference system, and K nearest neighbor models based on collection of 207 laboratory tests, are investigated for compressive strength prediction of concrete at high temperature. In addition for each model, two different sets of input variables are examined: a complete set and a parsimonious set of involved variables. The results obtained are compared with each other and also to the equations of NIST Technical Note standard and demonstrate the suitability of using the data driven models to predict the compressive strength at high temperature. In addition, the results show employing the parsimonious set of input variables is sufficient for the data driven models to make satisfactory results.

Due to the finite sources of natural aggregates, recycled concrete aggregates can be used as a suitable and economic material for the production of recycled concrete. Use of the recycled aggregates in concrete may undermine some mechanical properties of concrete. The aim of this study is to determine an optimum mix design made with the recycled aggregates, so that the environmental and economic aspects and also the fresh and hardened properties of concrete are taken into account. In the first phase of this research, different recycled concrete mixes with different levels of replacement of the recycled fine and recycled coarse aggregates are tested respectively instead of the natural fine and natural coarse aggregates. The results show the replacement of fine aggregates at the level of 30% has no significant negative effect on the expected common fresh and hardened properties of concrete such as fresh density, workability and compressive strength, and so this mix design can be selected as the optimum recycled mix design. In order to extend the application of the recycled concrete for the structures that tensile and flexural strengths may be demanded too, in the second phase of the research, the effect of the adding of polypropylene and steel fibers on the tensile and flexural strengths of the optimum mix design of the recycled concrete is investigated. The results show that the use of the fibers as the partial replacement of cement has no significant effect on the fresh properties of concrete, while effectively increases the tensile and flexural strengths of the recycled concrete and improve crack resistance and brittle behavior of the concrete when failure occurs.

The use of additives and their proportioning in concrete to achieve better performance is one of the challenging issues in concrete technology. In this study, the influence of different percentages of recycled fine and coarse aggregates on rheological, mechanical, and microstructural properties of self-compacting concrete (SCC) is investigated. For this purpose, fine and coarse natural aggregates (FNA and CNA) were replaced with fine and coarse recycled concrete aggregates (FRCA and CRCA), respectively, at replacement rates of 25%, 50%, 75% and 100% (by weight). The results showed that in order to meet the rheological requirements of SCC, the quantity of FRCA or CRCA should be limited to 25%. Furthermore, the mechanical properties showed a decreasing trend with FRCA and CRCA content, however, all mixes satisfied the strength requirement of structural concrete (28-day compressive strength of 21 MPa) even at 100% CRCA or FRCA content. In order to enhance the mechanical properties of mixes, 5%, 7.5%, and 10% (by weight) of micro-silica was used as a partial replacement of cement. The results showed that the mix with 25% recycled aggregates and 5% micro-silica presented comparable performance to that of the plain concrete, while using a higher micro-silica content negatively affected the rheological properties. In addition, the effect of recycled aggregate and micro-silica on the microstructure of mixes were examined.

Purpose This study aims to propose the adoption of artificial neural network (ANN)-based prediction intervals (PIs) to give more reliable prediction of labour productivity using historical data. Design/methodology/approach Using the proposed PI method, various sources of uncertainty affecting predictions can be accounted for, and a PI is proposed instead of a less reliable single-point estimate. The proposed PI consists of a lower and upper bound in which the realization of the predicted variable, namely, labour productivity, is anticipated to fall with a defined probability and represented in terms of a confidence level (CL). Findings The proposed PI method is implemented on a case study project to predict labour productivity. The quality of the generated PIs for the labour productivity is investigated at three confidence levels. The results show that the proposed method can predict the value of labour productivity efficiently. Practical implications This study is the first attempt in construction management to undertake a shift from deterministic point predictions to interval forecasts to improve the reliability of predictions. The proposed PI method will help project managers obtain accurate and credible predictions of labour productivity using historical data. With a better understanding of future outcomes, project managers can adopt appropriate improvement strategies to enhance labour productivity before commencing a project. Originality/value Point predictions provided by traditional deterministic ANN-based forecasting methodologies may be unreliable due to the different sources of uncertainty affecting predictions. The current study proposes ANN-based PIs as an alternative and robust tool to give a more reliable prediction of labour productivity using historical data. Using the proposed method, various sources of uncertainty affecting the predictions are accounted for, and a PI is proposed instead of a less reliable single point estimate.

Optimization of truss structures including topology, shape and size optimization were investigated by different researchers in the previous years. The aim of this study is discrete and continuous size optimization of two-dimensional truss structures with the fixed topology and the shape. For this purpose, the section area of the members are considered as the decision variables and the weight minimization as the objective function. The constraints are the member stresses and the node displacements which should be limited at the allowable ranges for each case. In this study, Genetic Algorithm and Particle Swarm Optimization algorithm are used for truss optimization. To analyse and determine the stresses and displacements, OpenSees software is used and linked with the codes of Genetic Algorithm and Particle Swarm Optimization algorithm provided in the MATLAB software environment. In this study, the optimization of four two-dimensional trusses including the Six-node, 10-member truss, the Eight-node, 15-member truss, the Nine-node, 17-member truss and the Twenty-node, 45-member truss under different loadings derived from the literature are done by the Genetic Algorithm and Particle Swarm Optimization algorithm and the results are compared with those of the other researchers. The comparisons show the outputs of the Genetic Algorithm are the most generally economical among the different studies for the discrete size cases while for the continuous size cases, the outputs of the Particle Swarm Optimization algorithm are the most economical.

Concrete corrosion is one of the most significant failure mechanisms of sewer pipes, and can reduce the sewer service life significantly. To facilitate the management and maintenance of sewers, it is essential to obtain reliable prediction of the expected service life of sewers, especially if that is based on limited environmental conditions. Recently, a long-term study was performed to identify the controlling factors of concrete sewer corrosion using well-controlled laboratory-scale corrosion chambers to vary levels of H2S concentration, relative humidity, temperature and in-sewer location. Using the results of the long-term study, three different data-driven models, i.e. multiple linear regression (MLR), artificial neural network (ANN), and adaptive neuro fuzzy inference system (ANFIS), as well as the interaction between environmental parameters, were assessed for predicting the corrosion initiation time (ti) and corrosion rate (r). This was performed using the sewer environmental factors as the input under 12 different scenarios after allowing for an initiation corrosion period. ANN and ANFIS models showed better performance than MLR models, with or without considering the interactions between environmental factors. With the limited input data available, it was observed that ti prediction by these models is quite sensitive, however, they are more robust for predicting r as long as the H2S concentration is available. Using the H2S concentration as a single input, all three data driven models can reasonably predict the sewer service life.

Evaluating the in situ concrete compressive strength by means of cores cut from hardened concrete is acknowledged as the most ordinary method, however, it is very difficult to predict the compressive strength of concrete since it is affected by many factors such as different mix designs, methods of mixing, curing conditions, compaction, etc. In this paper, considering the experimental results, three different models of multiple linear regression model (MLR), artificial neural network (ANN), and adaptive neuro-fuzzy inference system (ANFIS) are established, trained, and tested within the Matlab programming environment for predicting the 28 days compressive strength of concrete with 173 different mix designs. Finally, these three models are compared with each other and resulted in the fact that ANN and ANFIS models enables us to reliably evaluate the compressive strength of concrete with different mix designs, however, multiple linear regression model is not feasible enough in this area because of nonlinear relationship between the concrete mix parameters. Finally, the sensitivity analysis (SA) for two different sets of parameters on the concrete compressive strength prediction are carried out.

In this research, recycled concrete aggregates are used instead of natural aggregates in concrete and then the recycled concrete is reinforced by fibers. The aim of this study is to investigate the effect of size and also the amount of the recycled concrete aggregates on the mechanical properties of concrete and then investigating the effect of addition of polypropylene and steel fibers on improving the mechanical properties of concrete. The results showed that compressive strength of concrete decreased by increasing replacement ratio of recycled aggregates. Also, the replacement of fine aggregate at the level of 30 percent has not significant effect on the compressive strength of concrete and so this mix can be selected as the optimum recycled mix design. In addition, the use of the fiber improves considerably the tensile and flexural strength of the recycled concrete so that the steel fiber has more positive effects on the tensile and flexural strength in comparison to the polypropylene fiber. In addition, when the concrete reinforced by the fiber is broken, the concrete is not disrupted and the fiber has an important role to reduce the growth of cracks after break. In this research, recycled concrete aggregates are used instead of natural aggregates in concrete and then the recycled concrete is reinforced by fibers. The aim of this study is to investigate the effect of size and also the amount of the recycled concrete aggregates on the mechanical properties of concrete and then investigating the effect of addition of polypropylene and steel fibers on improving the mechanical properties of concrete. The results showed that compressive strength of concrete decreased by increasing replacement ratio of recycled aggregates.Also, the replacement of fine aggregate at the level of 30 percent has not significant effect on the compressive strength of concrete and so this mix design can be selected as the optimum recycled mix design. In addition, the use of the fibers improve considerably the tensile and flexural strength of the recycled concrete so that the steel fiber has more positive effects on the tensile and flexural strength in comparison to the polypropylene fiber. Also, when the concrete reinforced by the fiber is broken, the concrete is not disrupted and the fiber has an important role to reduce the growth of cracks after break.

In recent years, numerous experimental tests were done on the concrete beams reinforced with the fiber-reinforced polymer (FRP). In this way, some equations were proposed to estimate the shear strength of the beams reinforced with FRP. The aim of this study is to explore the feasibility of using a feed-forward artificial neural network (ANN) model to predict the ultimate shear strength of the beams strengthened with FRP composites. For this purpose, a database consists of 304 reinforced FRP concrete beams have been collected from the available articles on the analysis of shear behavior of these beams. The inputs to the ANN model consists of the 11 variables including the geometric dimensions of the section, steel reinforcement amount, FRP amount and the properties of the concrete, steel reinforcement and FRP materials while the output variable is the shear strength of the FRP beam. To assess the performance of the ANN model for estimating the shear strength of the reinforced beams, the outputs of the ANN are compared to those of equations of the Iranian code (Publication No. 345) and the American code (ACI 440). The comparisons between the outputs of Iran and American regulations with those of the proposed model indicates that the predictive power of this model is much better than the experimental codes. Specifically, for under study data, mean absolute relative error (MARE) criteria is 13%, 34% and 39% for the ANN model, the American and the Iranian codes, respectively.

Decision making on buildings after the earthquake have always been a great concern of scientists. Safety concerns, possibility of using the building, repairing the building, and the rate of damage are some of the most vital factors that needs to be paid attention in immediate decision makings of the buildings. In order to determine the level of damage in the buildings, the maximum displacement of stories is one of the most important parameter that needs to be investigated. In this paper, a concrete frame with shear wall containing 4-stories and 4-bays has been designed for acceleration records of 0.1g to 1.5g and the rate of damage is determined. The total of 450 data with 6 input variables and one output variable is produced. The input parameters are defined as frequency, Vs, Richter, the distance from the earthquake epicentre (DEE), PGA, and acceleration, and the output parameter is defined as drift. With respect to this data set, three different data-driven models, i.e. Artificial Neural Network (ANN), Adaptive Neuro-Fuzzy Inference System (ANFIS), and Multiple Linear Regression Model (MLR) are used to predict the displacements. Results indicate that Both the ANN and ANFIS model show great accuracies in estimating the displacements in concrete frame with shear wall. On the other hand, MLR model did not show acceptable accuracy in the same estimation purposes. Finally, the sensitivity analysis was performed on the data set and it was observed that the accuracy of the predictions highly depends on the number of input parameters. In other words, increasing the number of input parameters would result in the increase in the accuracy of the final prediction results.

Ultrasonic pulse velocity (UPV) test method is used in this study for evaluating the compressive strength of concrete. A series of UPV tests were performed to evaluate the 28-day compressive strength of concrete and examine the effect of concrete mixture parameters on the UPV of concrete. It was found that concrete with higher 28-day compressive strength gives higher UPV and that an exponential relationship exists between the UPV and 28-day compressive strength of concrete. The results showed that the aggregate size has a significant effect on the strength of concrete. Concrete with larger aggregate size was found to give lower UPV and compressive strength. UPV results also indicated that the UPV and compressive strength of concrete consistently decrease with increase in water-cement ratio of concrete. The effect of using microsilica (Silica fume) in concrete is also studied. It was found that as the microsilka to cement ratio increases in concrete, the UPV and compressive strength of concrete increase. The effects of the ingredient materials on UPV were analyzed and potential mechanisms were proposed. To make the results applicable, the artificial neural network (ANN) method was used to predict the compressive strength of concrete based on the evaluated concrete mix parameters and ultrasonic pulse velocity. The ANN analysis demonstrated high reliability in predicting the compressive strength values of concrete.

Regardless of extensive researches on hydrologic forecasting models, the issue of updating the outputs from forecasting models has remained a main challenge. Most of the existing output updating methods are mainly based on the presence of persistence in the errors. This paper presents an alternative approach to updating the outputs from forecasting models in order to produce more accurate forecast results. The approach uses the concept of the similarity in errors for error prediction. The K nearest neighbor (KNN) algorithm is employed as a similarity-based error prediction model and improvements are made by new data, and two other forms of the KNN are developed in this study. The KNN models are applied for the error prediction of flow forecasting models in two catchments and the updated flows are compared to those of persistence-based methods such as autoregressive (AR) and artificial neural network (ANN) models. The results show that the similarity-based error prediction models can be recognized as an efficient alternative for real-time inflow forecasting, especially where the persistence in the error series of flow forecasting model is relatively low.

The main objective of this study is to present a simulation‐optimization approach methodology for short‐term operation of multi‐objective reservoirs during emergency conditions where risk of flooding and overflow from damaged spillways is significant. The paper considers flood risk management through enhancing dam safety as well as long‐term objectives of the reservoir for hydropower generation and irrigation water supply simultaneously. In particular, real‐time operation of Abbaspour reservoir during an emergency condition is considered where all flood discharging facilities are out of order for repairing the damaged parts. The proposed methodology combines a short‐term and a long‐term operation strategies to minimize risk of overflow for any possible incoming floods while preserving the long‐term objectives of the reservoir. An efficient daily flow forecasting scheme provides the inflow data for the short‐term (daily) operation strategy in a deterministic environment which follows the long‐term operation policy. A multi‐objective stochastic dynamic programming model addresses the long‐term operation policy by solving the weekly operation model. The outcomes of the model application to the case study have illustrated its capabilities in developing a compromise between the risk of dam overflow and its long‐term objectives.

Instance based learning (IBL) algorithms are a common choice among data driven algorithms for inflow forecasting. They are based on the similarity principle and prediction is made by the finite number of similar neighbors. In this sense, the similarity of a query instance is estimated according to the closeness of its feature vector with those of data available in calibration data. As the selected attributes in the feature vector are determined overall on calibration data, there may be some data points whose outputs do not follow the considered attributes. In fact, output values of these inconsistent data points may be a function of some other attributes which were not considered. Therefore, for some query instances, the inconsistent points may be appeared as the neighbors while they may not really be neighbor to the query instance. They can deteriorate forecasting results especially if they are very close to the query instance with the current similarity definition. In this study a clustered K nearest neighbor (CKNN) algorithm is introduced which can capture these inconsistent data points. Similar to the inconsistent data points, CKNN can be also robust against noisy data. The proposed algorithm was shown to be effective for a synthetic linear data set corrupted by noise. In addition, the utility of the algorithm was demonstrated for daily inflow forecasting of the Karoon1 reservoir located in Iran.

Multiobjective reservoir operations are generally complex, as they are often associated with a large quantity of uncertain factors in combination with noncommensurable objectives. In this study, fuzzy-state stochastic dynamic programming (FSDP) and multicriteria decision-making (MCDM) are integrated to derive operating rules for a single multiobjective reservoir operation problem. The model addresses uncertainties due to randomness in inflows and imprecision in variables' discretization and objectives. The FSDP model takes into account uncertainties due to the random nature of inflows and imprecision due to variable discretization. Imprecise and noncommensurable objectives are quantified by a set of subjective criteria, the aggregation of which is performed through an MCDM model, by which possible decisions at every stage of the FSDP model are evaluated and compared. The proposed approach is then employed in deriving operating rules for the Karoon 1 reservoir in south-west Iran and the rules are tested and evaluated through simulation. Results show the model's capability in handling different kinds of uncertainties involved in real reservoir operation problems.

Reconstruction and/or modification of an already existing fuzzy model with new data may improve system performances. As new data become available, adjusting the existing fuzzy rule-based model may present a challenging alternative to full model reconstruction. In this paper a fuzzy rule-based control model using a Takagi–Sugeno fuzzy system is presented and a model modification algorithm is developed which improves the performance of the initial model as new data become available. Proposed approach is applied to a flood flow forecasting case example and the results are compared with those forecasted using initially available and reconstructed models. Results show that the modified model outperforms the initial FRB model. Reconstructed model performs slightly better than the modified model; however, the reconstruction may not be justified in a real time flood forecasting system, considering the limitations on the available lead time.