Search

Purpose Accurate and rapid tracking and counting of building materials are crucial in managing on-site construction processes and evaluating their progress. Such processes are typically conducted by visual inspection, making them time-consuming and error prone. This paper aims to propose a video-based deep-learning approach to the automated detection and counting of building materials. Design/methodology/approach A framework for accurately counting building materials at indoor construction sites with low light levels was developed using state-of-the-art deep learning methods. An existing object-detection model, the You Only Look Once version 4 (YOLO v4) algorithm, was adapted to achieve rapid convergence and accurate detection of materials and site operatives. Then, DenseNet was deployed to recognise these objects. Finally, a material-counting module based on morphology operations and the Hough transform was applied to automatically count stacks of building materials. Findings The proposed approach was tested by counting site operatives and stacks of elevated floor tiles in video footage from a real indoor construction site. The proposed YOLO v4 object-detection system provided higher average accuracy within a shorter time than the traditional YOLO v4 approach. Originality/value The proposed framework makes it feasible to separately monitor stockpiled, installed and waste materials in low-light construction environments. The improved YOLO v4 detection method is superior to the current YOLO v4 approach and advances the existing object detection algorithm. This framework can potentially reduce the time required to track construction progress and count materials, thereby increasing the efficiency of work-in-progress evaluation. It also exhibits great potential for developing a more reliable system for monitoring construction materials and activities.

Purpose Images taken from construction site interiors often suffer from low illumination and poor natural colors, which restrict their application for high-level site management purposes. The state-of-the-art low-light image enhancement method provides promising image enhancement results. However, they generally require a longer execution time to complete the enhancement. This study aims to develop a refined image enhancement approach to improve execution efficiency and performance accuracy. Design/methodology/approach To develop the refined illumination enhancement algorithm named enhanced illumination quality (EIQ), a quadratic expression was first added to the initial illumination map. Subsequently, an adjusted weight matrix was added to improve the smoothness of the illumination map. A coordinated descent optimization algorithm was then applied to minimize the processing time. Gamma correction was also applied to further enhance the illumination map. Finally, a frame comparing and averaging method was used to identify interior site progress. Findings The proposed refined approach took around 4.36–4.52 s to achieve the expected results while outperforming the current low-light image enhancement method. EIQ demonstrated a lower lightness-order error and provided higher object resolution in enhanced images. EIQ also has a higher structural similarity index and peak-signal-to-noise ratio, which indicated better image reconstruction performance. Originality/value The proposed approach provides an alternative to shorten the execution time, improve equalization of the illumination map and provide a better image reconstruction. The approach could be applied to low-light video enhancement tasks and other dark or poor jobsite images for object detection processes.

PurposeMeasuring onsite productivity has been a substance of debate in the construction industry, mainly due to concerns about accuracy, repeatability and unbiasedness. Such characteristics are central to demonstrate construction speed that can be achieved through adopting new prefabricated systems. Existing productivity measurement methods, however, cannot cost-effectively provide solid and replicable evidence of prefabrication benefits. This research proposes a low-cost automated method for measuring onsite installation productivity of prefabricated systems.Design/methodology/approachFirstly, the captured ultra-wide footages are undistorted by extracting the curvature contours and performing a developed meta-heuristic algorithm to straighten these contours. Then a preprocessing algorithm is developed that could automatically detect and remove the noises caused by vibrations and movements. Because this study aims to accurately measure the productivity the noise free images are double checked in a specific time window to make sure that even a tiny error, which have not been detected in the previous steps, will not been amplified through the process. In the next step, the existing side view provided by the camera is converted to a top view by using a spatial transformation method. Finally, the processed images are compared with the site drawings in order to detect the construction process over time and report the measured productivity.FindingsThe developed algorithms perform nearly real-time productivity computations through exact matching of actual installation process and digital design layout. The accuracy and noninterpretive use of the proposed method is demonstrated in construction of a multistorey cross-laminated timber building.Originality/valueThis study uses footages of an already installed surveillance camera where the camera's features are unknown and then image processing algorithms are deployed to retrieve accurate installation quantities and cycle times. The algorithms are almost generalized and versatile to be adjusted to measure installation productivity of other prefabricated building systems.

Many studies show that the human energy-related behaviors have a significant impact on the return of Energy Efficiency Programs (EEPs). However, studies that aimed at increasing the energy savings from the EEPs are still limited. In this paper, a Genetic Agent-Based (GAB) framework has been proposed to enhance the return of a typical EEP by simulating social network and energy behavior attributes and finding the best participants among a target community. Several attributes are considered for creating the agent-based model of households and numerically representing their interactions with the EEP or within their social network. The improvement of the EEP using the GAB framework is tested on a social network consisting of 56 households. The simulation results show that by accurately selecting participants using the presented framework, the amount of energy saving could increase up to ten times. This ultimately indicates the considerable impact of the social network on the EEP performance. In other words, to have an efficient EEP in the long term, the social network attributes such as network degree and strength of connections should be also considered in decision-making along with the energy-related attributes.

Appropriate planning and strategies in Municipal Solid Waste (MSW) management could significantly decrease environmental and economic costs imposing on both management authorities and local communi...

Collecting effective data is a fundamental step in developing transport networks and related research. Social media have become an emerging source of data for traffic analyses. In this paper, we demonstrate that the function of a city influences the utility of social media data in travel demand models by generating models for eight US cities with different functions. Data from Twitter and Foursquare, as well as other socio-demographic information, are considered as independent variables in Origin-Destination trip regression models generated via a Random Forest regression technique. Model performance with and without use of social media data are compared via 10-fold cross-validation. The results indicate that the accuracy of the models for all eight cities improved when independent variables based on social media data were included. The performance was most improved in metropolitan areas, followed by rural and tourist areas. Inspired by this finding, we conclude that the city function influences the utility of social media data in travel demand models. Meanwhile, we create models based on trip purpose and transport mode to explore other factors that may impact the efficiency of applying social media data in transport research.

Political conduct is a normal occurrence in every organization in today's business environment. The political environment in the construction industry is particularly problematic because o...

© 2018 Elsevier B.V. The availability of resources such as workforce and materials at each level of a high-rise construction project just before the commencement of building tasks is a crucial issue that might have direct impacts on project progress. To avoid delays caused by lack of human resources and construction materials, a construction management team always tries to find a better way to facilitate supply chain process specifically for construction projects facing a significant number of simultaneous and repetitive tasks. The other challenge in a high-rise construction project is vertical transportation that requires special machinery e.g. cranes or lifts, and also, maximizing their utilities. In this paper, it is tried to automate vertical transportation planning process in high-rise construction projects by introducing a platform that handles the entire lifting process. This platform considers (i) tasks attributes (e.g. required resources, location and commencement time) from the project schedule, (ii) lifting system specifications (e.g. travel speed, weight, and volume capacity) and (iii) project geometrics (e.g. current height of the project). In details, the introduced platform provides an optimized daily delivery plan by developing a Mixed-Integer Programming (MIP) model that covers workforce and construction materials. In this paper, the proposed platform is also tested using field data obtained from a 34-story construction project in Mashhad, Iran. The model could find a solution with 0% optimality gap in approximately 1 h, which is an acceptable amount of computational cost for the problem. The results show how the introduced platform can assist the construction management team to efficiently handle the supply process within stories while avoiding delays caused by a lack of resources required for each task.

The Ready Mixed Concrete Dispatching Problem (RMCDP) is a generalised version of the Vehicle Routing Problem (VRP). In RMCDP however, given the particular properties of concrete, a set of additional constraints must be considered. The perishability of fresh material, for example, has to be contemplated, together with other technical issues, such as a truck can only serve a single customer and two or more customers cannot be supplied using a truck in a single trip. RMCDP can be modelled as a mixed integer problem (with soft time window) and it can be optimally solved for small and medium sized variants of the problem. However, the objective cost and similar constraints are usually calculated using spatial distances (Euclidian distance) and similar approximate travel times are used. In this paper, we propose a more realistic method for calculating travel times based on real travel distances obtained from Google Maps API. The accurate travel time information is used as input for the RMCDP, to provide...

In recent years, there has been an increasing incidence of failure of rock bolts due to stress corrosion cracking and localized corrosion attack in Australian underground coal mines. Unfortunately, prediction of the risk of failure from results obtained from laboratory testing is not necessarily reliable because it is difficult to properly simulate the mine environment. An alternative way of predicting failure is to apply machine learning methods to data obtained from underground mines. In this paper, support vector machines are built to predict failure of bolts in complex mine environments. Feature transformation and feature selection methods are applied to extract useful information from the original data. A dataset, which had continuous features and spatial data, was used to test the proposed model. The results showed that principal component analysis-based feature transformation provides reliable risk prediction.

Sanitary waste disposal and site selection for establishing landfills are challenging problems for environmental planners. This paper aims to take environmental, socio-economic, geological, geomorphological, hydrological and ecological factors into consideration to provide a decision support framework for landfill siting. Analytical hierarchy process (AHP) and Decision Making Trial and Evaluation Laboratory (DEMATEL) are coupled to develop an efficient multi-criteria decision-making method to be utilized in a Geographic Information System (GIS) environment for evaluating the suitability for landfill siting. As the first attempt to employ DEMATEL effectively in a landfill site selection problem, the proposed method is tested with landfill siting scenarios in New South Wales (NSW), Australia. Regional analysis is also performed to identify the potentially most suitable statistical divisions for landfill siting in NSW. The top two ranked zones covering 0.7% and 22% of the study area, respectively, are considered as the optimal areas for establishing landfills, while the bottom two ranked zones are not recommended for further consideration. Further detailed analysis is also conducted on the existing landfills, which shows that 1.0% and 37.0% of them are ranks 1 and 2, respectively. The scenario-based analysis implies that, among the contributing factors; geological and economic factors are highly important.

For many years, simultaneous evacuation of the entire facility was a common approach. The deficiencies of this approach were revealed after many cases of casualties and evacuation efficiency drop that were observed during some evacuations caused by crowd congestion. To tackle this issue, staged-evacuation process (also called phased-evacuation), which was mainly aimed to mitigate the congestions, was introduced. In this study, a bi-level framework consisted of a pedestrian simulation model (Social Force Model) and a metaheuristic optimization model (GA) was developed to automatically determine the optimum schedule for a staged-evacuation process. In this platform, metaheuristic algorithm iteratively searches for a dominant evacuation schedule, and SFM simulates the evacuees' attributes and evaluates the safety of the evacuation scenario which was proposed by the metaheuristic algorithm. Total Evacuation Time (TET), fire exposure threat, and congestion severity were also considered as the safety objectives which were tried to be minimized. An opportunity was also provided to cover the effects of many pedestrians’ exceptional behaviors by executing a high number of microscopic simulations during the metaheuristic iterations. Moreover, more reliable input materials were provided by the microscopic simulator for the optimization algorithm. In this regard, executing multiple simulations in every metaheuristic search demands significant computation efforts. To overcome this challenge, a parallel computing technique was hired which resulted in a considerable reduction in the computation time. Finally, an MCDM method was employed to select the best-satisfying evacuation scenario among the Pareto-frontier solutions. Implementing the proposed framework in a case study enabled us to evacuate the examined facility based on the achieved optimized schedule so that the congestion severity tremendously mitigated while fire exposure threat and TET remained compromisingly at their lowest.

Unmanned aerial vehicles (UAV) have made life easier in many ways, and their applications in civil practice are increasing rapidly. However, this benefit is not entirely risk-free, as unwanted accidents and incidents can cause serious harm and interrupt other aerial activities. In this paper, we investigate a dataset of UAV accidents and incidents in Australia and put up some precautionary exercises to reduce the risk of future events. To that end, univariate and bivariate distributions of past events are analysed, and the exploratory factor analysis technique is used to identify frequent accident and incident patterns. The findings show that equipment issues or/and lack of coordination between aerial activities are two of the accidents and incidents categories; therefore, necessitating regular safety inspections for UAVs and establishing an integrated monitoring system for aerial activities are expected to reduce the risk of accidents and incidents.

In the past few years, the social science literature has shown significance attention to extracting information from social media to track and analyse human movements. In this paper the transportation aspect of social media is investigated and reviewed. A detailed discussion is provided about how social media data from different sources can be used to indirectly and with minimal cost extract travel attributes such as trip purpose, mode of transport, activity duration and destination choice, as well as land use variables such as home, job and school location and socio-demographic attributes including gender, age and income. The evolution of the field of transport and travel behaviour around applications of social media over the last few years is studied. Further, this paper presents results of a qualitative survey from travel demand modelling experts around the world on applicability of social media data for modelling daily travel behaviour. The result of the survey reveals positive view of the experts about usefulness of such data sources.

Duration of concrete pouring task is typically considered from the arrival time of the first Ready Mixed Concrete (RMC) truck until the end of the pouring process of the last truck. In practice, the concrete pouring duration does not only depend on site features. In fact, the duration is also affected by other parameters such as supply process, location of the project and traffic conditions, particularly in cities with heavy traffic. This paper investigates impacts of supply chain parameters on predicting concrete pouring duration that have been typically excluded from such analyses. Unlike other studies which are limited to construction site parameters in predicting concrete pouring duration, this study not only considers construction site factors at a general level but also investigates the impact of supply related parameters. To test the effectiveness of considered variables a field dataset of an active RMC in Adelaide, Australia with four batch plants and around 40 trucks is used. The dataset covers a period of a month which includes 2658 deliveries to > 500 locations. In terms of the modeling practice of this study, first a preliminary linear regression is developed and then it is modified to satisfy crucial assumptions of heteroscedasticity and residuals normality. Finally, a hazard-based model where the assumption of residual normality is relaxed, is developed. The results show that severe bias occurs when assumptions associated with linear regression are overlooked. Moreover, in the developed models the supply parameters are found to have significant impacts on concrete pouring duration.

Ready Mixed Concrete (RMC) suffers from a lack of practical solutions for automatic resource allocation. Under these circumstances, RMC dispatching systems are mostly handled by experts. This paper attempts to introduce a machine learning based method to automatically match experts' decisions in RMC. For this purpose, seven machine learning techniques with their boosted algorithms were selected. A set of attributes was extracted from the collected field data. Eleven metrics were used to assess the performance of the selected techniques using different approaches. Due to concerns about randomness, significant testing was performed to assist in finding the best algorithm for this purpose. Results show that Random-Forest with 85% accuracy outperforms the other selected techniques. One of the most interesting achieved results is related to the computing time. The results show that all the selected algorithms can solve large-scale depot allocations with a very short computing time. This is possibly because a model built by a machine learning algorithm only needs to be tested with new instances, which does not need an extensive computation effort. This provides us with a chance to move toward automation in Ready Mixed Concrete Dispatching Problems (RMCDPs), especially for those RMCs with dynamic environments where resource allocation might need to be quickly recalculated during the RMC process due to changes in the system.

Purpose The purpose of this paper is to predict the concrete pouring production rate by considering both construction and supply parameters, and by using a more stable learning method. Design/methodology/approach Unlike similar approaches, this paper considers not only construction site parameters, but also supply chain parameters. Machine learner fusion-regression (MLF-R) is used to predict the production rate of concrete pouring tasks. Findings MLF-R is used on a field database including 2,600 deliveries to 507 different locations. The proposed data set and the results are compared with ANN-Gaussian, ANN-Sigmoid and Adaboost.R2 (ANN-Gaussian). The results show better performance of MLF-R obtaining the least root mean square error (RMSE) compared with other methods. Moreover, the RMSEs derived from the predictions by MLF-R in some trials had the least standard deviation, indicating the stability of this approach among similar used approaches. Practical implications The size of the database used in this study is much larger than the size of databases used in previous studies. It helps authors draw their conclusions more confidently and introduce more generalised models that can be used in the ready-mixed concrete industry. Originality/value Introducing a more stable learning method for predicting the concrete pouring production rate helps not only construction parameters, but also traffic and supply chain parameters.

Building information modeling (BIM) has emerged as a pervasive and ubiquitous tool in the fields of construction and engineering. With BIM’s rise, so too has there been an explosion of rese...

The pavement performance model is a basic part of the pavement management system. The prediction accuracy of the model depends on the number of effective variables and the type of mathematical method that is used for modeling the pavement performance. In this paper, the capability of the support vector machine (SVM) method is analyzed for predicting the future of the pavement condition. Five kernel types of SVM algorithm are formed and nine input variables of the proposed models are extracted from the range of effective variables on the pavement condition. The international roughness index is used as the pavement performance index. The results show that the Pearson VII Universal kernel can accurately predict pavement performance in its life cycle.

Purpose – The purpose of this paper is to study the implementation of machine learning (ML) techniques in order to automatically measure the feasibility of performing ready mixed concrete (RMC) dispatching jobs. Design/methodology/approach – Six ML techniques were selected and tested on data that was extracted from a developed simulation model and answered by a human expert. Findings – The results show that the performance of most of selected algorithms were the same and achieved an accuracy of around 80 per cent in terms of accuracy for the examined cases. Practical implications – This approach can be applied in practice to match experts’ decisions. Originality/value – In this paper the feasibility of handling complex concrete delivery problems by ML techniques is studied. Currently, most of the concrete mixing process is done by machines. However, RMC dispatching still relies on human resources to complete many tasks. In this paper the authors are addressing to reconstruct experts’ decisions as only practical solution.

Concrete delivery dispatching suffers from a lack of practical solutions and therefore, in the absence of automatic solutions, experts are hired to handle this task. In addition, the concrete delivery dispatching problem can be modelled mathematically but it can only solve up to medium sizes of this problem within a practical time. This paper attempts to answer the question of how much we can rely on experts' decisions. First, the concrete delivery problem is presented. Second, a benchmark for the problem is achieved; two heuristic methods are used for those instances that their exact solutions are not available. Finally, the experts' decisions are compared with the obtained benchmarks to assess the optimality gap of the experts. A field dataset which belongs to an active Ready Mixed Concrete (RMC) is used to evaluate the proposed idea. The results show that experts' decisions are near to optimum, with an average accuracy of 90%. However, after comparing individual decisions between optimisation models and the experts' decisions, we can conclude that optimisation models only try to achieve the lowest cost, while the expert prefers a more stable dispatching system at slightly higher cost.

The logistics and planning problem of delivering ready mixed concrete (RMC) to a set of demand customers from multiple depots is addressed. The RMC dispatching problem (RMCDP) is closely related to the vehicle routing problem, with the difference that a truck may visit demand nodes in the RMCDP more than once. This class of routing problems can be represented by using mixed-integer programming (MIP) and is known to be NP-hard. Solving RMC delivery problems is often achieved through heuristics and metaheuristic-based methods as exact solution approaches are often unable to find optimal solutions efficiently, in particular when multiple depots are represented in the model. Although a variety of methods are available to solve MIP models, in this paper an attempt is made to solve the RMCDP by using a Lagrangian relaxation technique. Namely, a solution algorithm based on Lagrangian relaxation is derived to reduce the complexity of the initial MIP model and show that the proposed relaxation is able to provide promising computation results on a realistic data set representative of an active RMCDP in the region of Adelaide, Australia.

Being able to precisely predict the duration of concrete operations can help construction managers to organize sites and machineries more efficiently, especially when there is limited space for equipment on site. Currently there is no theoretical method for estimating the duration of the concrete pouring process. Normally, the maximum capacity of pumping facilities on construction sites is not used, and concrete pumps are idle for a considerable time as a result of the arrival of concrete trucks being delayed. In the light of this issue, this paper considers the supply chain parameters of Ready Mixed Concrete (RMC) as a means of solving this problem. Artificial Neural Network (ANN) is hired for modelling/predicting the productivity of a concrete operation. The proposed model is tested with a real database of an RMC in the Sydney metropolitan area that has 17 depots and around 200 trucks. Results show that there is an improvement in the achieved results when these are compared to the results of relevant studies that only considered the construction parameters for predicting the productivity of concrete operations

© 2019 Elsevier B.V. Timely delivery of materials and laborers to workface is critical to the smooth progression of a task on the construction site. This subject is of particular importance in high-rise building projects where the vertical transportation may turn into a bottleneck in the delivery process, if not precisely planned. In such contexts, lift cars represent a contested resource for movement of people and materials and any delayed lifting leads to disrupted tasks on the building floors. The available literature predominantly relies on predefined or rule-based scenarios in addressing the lifting problem which may result in suboptimal solutions to the operation of the lifts. Therefore, this research aims to develop a holistic mathematical model which identifies the optimum operational plan for the entire lift systems in high rise construction. The model has a mixed integer nature incorporating characteristics and constraints of lift cars, construction resources and activities, and site conditions. The optimality is targeted through minimum travel time both at the whole scale and for a single lift car. To improve generalizability, the model has been adapted to variant lift optimization problems under different construction scenarios. It is then integrated with the zoning problem in order to determine optimal zoning configuration for a given lift system. The model is tested on a hypothetical case replicating complexities and constraints of real-world situations. The results indicate that the model can efficiently find an optimal solution for the lift systems which imposes minimum vertical travel time to the project.

One of the major causes of delays in construction sites belongs to the delays in delivering materials. This paper introduces the integrated Building Information Modelling (BIM) model combined with a supply chain and a controlling system, along with its efficiency for alleviating the negative effects of delays caused by a lack of materials on construction sites. So, firstly the concept of feed-forward control is discussed as a prevention feature in a monitoring system. Then the integrating BIM model with a supply chain is introduced. Finally this case study has shown how the enhanced BIM model can incorporate the feed-forward control in order to decrease the risks of delays caused by a lack of materials in construction. Moreover, this case study might be used as a decision support system for managers to prevent a construction site from future delays by providing valuable information about the required materials in advanced

The earthmoving process is usually handled by a combination of multi-functional excavators and delivery trucks. According to the principles of excavation, a specific machine is selected to dig the earth. Sometimes, due to the difficulties in the process, the digging process is split between different machines with different capabilities. This paper aims to introduce the concept of dynamic equipment for allocating different tasks in an operation to a machine. This concept is discussed and modelled via a discrete-event simulation method and is tested in a real earthmoving operation from three different perspectives: time, cost and emission. The results from the case study show that the use of dynamic equipment will lead to a decrease in cost and emissions and an increase in productivity.

Reliability and risk analysis concerns analyzing and predicting the state transition of engineered systems given historic information. With an increasing volume of available data, conventional machine learning algorithms may fail to capture hidden patterns behind the reliability engineering problems due to the curse of dimensionality. Autoencoders, a class of deep learning frameworks, have been reviewed in this study. They are a powerful tool in extracting hidden representations and producing a robust reconstruction for further predicting tasks. We have developed autoencoders and applied them on four data sets. Results showed that autoencoders can not only classify the states at a good accuracy, but also help to discover the failure mechanism.