Search results

This study aims to (1) develop an artificial intelligence (AI)-based model to accurately forecast rebar prices and (2) propose procurement strategies to reduce the subjectivity involved in rebar price trend forecasting and minimize procurement costs for construction project general contractors.

Correlation analysis was used to identify the key factors influencing changes in rebar prices over time. An AI-based inference model, symbiotic bidirectional gated recurrent unit (SBiGRU), was developed for rebar price forecasting. The performance of SBiGRU was compared with other AI techniques, and procurement strategies based on the SBiGRU model were proposed.

The SBiGRU model outperformed the other AI techniques in terms of rebar price forecasting accuracy. The proposed rebar price forecasting model (RPFM) and procurement patterns, which integrate inventory management principles and rebar price forecasts, were demonstrated to effectively optimize procurement costs, realizing a remarkable 6.13% reduction in procurement expenses compared to the conventional monthly procurement approach.

The accuracy of AI models may be impacted by disparities in the data used for model training. Future research should explore approaches incorporating price predictions and order factors.

This study significantly extends the bounds of traditional rebar price prediction by integrating AI-driven forecasting with inventory management principles, highlighting the potential of AI-based models to improve construction industry procurement practices, reduce related risks and costs, optimize project operations and maximize project outcomes.

Work interruptions (WIs) due to social media are becoming more and more common in the daily lives of organizations. However, the relationship between WI and work performance of employees is still unclear. This study aims to investigate the effects of WIs due to social media on employees' work performance in terms of different mechanisms; it also considers the moderating role of social media usage.

Using the jobs demands-resource (JD-R) model, this paper proposes a research model to investigate the effects of WIs on employee work performance from the perspective of the enabling mechanism and burden mechanism. Structural equation modeling (SEM) was used to analyze the data of 444 employees.

The results show that (1) with regard to the enabling mechanism path, WI has a positive effect on employees' sense of belonging, which further has a positive effect on employees' work performance; (2) with regard to the burden mechanism path, WI has a positive effect on employees' interruption overload; however, the effect of employee interruption overload on employees' work performance is not significant, and (3) social media used for either work or social purposes can strengthen the relationship between WI and interruption overload, while social media used for work-related purposes can reduce the relationship between WI and a sense of belonging.

First, this paper contributes to the WI literature by clarifying how WI affects employees' work performance through different mechanisms, namely the enabling mechanism and the burden mechanism. Second, this paper contributes to the WI literature by revealing a boundary condition, namely social media use, between WI and a sense of belonging and between WI and employees' interruption overload.

This paper aims to conduct work to obtain high-quality brazed joint of YAG ceramic and kovar alloy.

Wetting and spreading behavior of AgCuTi filler alloy on YAG ceramic and kovar alloy under vacuum (2∼3 × 10–4 Pa) and argon conditions was investigated and compared. Then, YAG ceramic was brazed to kovar alloy under a high vacuum of 2∼3 × 10–4 Pa; the influence of holding time on the interface structure of the joint was investigated.

The wettability of AgCuTi on YAG is poor in the argon atmosphere, the high oxygen content in the reaction layer hinders the formation of the TiY2O5 reaction layer, thereby impeding the wetting of AgCuTi on YAG; in the vacuum, a contact angle (?=16.6°) is obtained by wetting AgCuTi filler alloy on the YAG substrate; the microstructure of the YAG/AgCuTi/kovar brazed joint is characterized to be YAG/Y2O3/(Fe, Ni)Ti/Ag(s, s) + Cu(s, s)/Fe2Ti + Ni3Ti/Fe2Ti/kovar; at 870 °C for the holding time of 10 min, a (Fe, Ni) Ti layer of approximately 1.8 µm is formed on the YAG side.

Wetting and spreading behavior of the brazing filler alloy under different conditions and the influence of the holding time on the interface microstructure of the joint were studied to provide references for obtaining high-quality brazed joints.

Laser hot wire additive manufacturing (LHWAM) is a newer technology within the space of large-scale directed energy deposition (DED) additive manufacturing (AM) processes. This study aims to map known AM flaw types such as lack of fusion and keyholing, as well as a dripping flaw unique to hot wire processes, across process parameter space using a small number of single-track experiments.

A semianalytical model was calibrated using a small initial set of experimental data. Lack of fusion and keyholing flaws were mapped across process space using existing models. The dripping flaw was modeled via analytical methods calibrated with experimental data, and then mapped across processing space. Further experimental data beyond the small initial set was used to evaluate the accuracy of the process maps developed. A website and executable were deployed to users of the process for convenient rapid process parameter selection.

With the process maps generated during this work, users can easily and rapidly generate desirable parameter sets for a range of conditions, enabling the intelligent utilization of the entire stable processing regime.

The methodology developed can be applied to other LHWAM machines or DED processes to rapidly and inexpensively generate a systematic understanding of processing space for build planning.

LHWAM shows advantages over other large-scale DED processes, but a systematic physically informed study of the key flaw regions across process space had not been conducted, limiting more widespread use of the process and creating a gap that this study fills.

This study aims to fabricate the reduced graphene oxide (rGO)/ethylene vinyl acetate copolymer (EVA) composite films with electric-driven two-way shape memory properties for deployable structures application. The effect of dicumyl peroxide (DCP) and rGO on the structure and properties of the rGO/EVA composite films were systematically investigated.

The rGO/EVA composite films were fabricated by melting blend and swelling-ultrasonication method, DCP and rGO were used the crosslinking agent and conductive filler, respectively.

The research results indicate that the two-way shape memory properties of rGO/EVA composite films were significantly improved with the increase of DCP content. The rGO endowed rGO/EVA composite films with excellent electric-driven reversible two-way shape memory and anti-ultraviolet aging properties. The sample rGO/EVA-9 can be heated above Tm within 8 s at a voltage of 35 V and can be heated above the Tm temperature within 12 s under near-infrared light (NIR). Under a constant stress of 0.07 MPa, the reversible strain of the sample rGO/EVA-9 was 8.96% and its electric-driven shape memory behavior maintained great regularity and stability.

The rGO/EVA composite films have potential application value in the field of deployable structures.

With the increase of DCP content, the two-way shape memory properties of rGO/EVA composite films were significantly improved, which effectively solved the problem that the shape memory properties of EVA matrix decreased caused by swelling. The rGO endowed rGO/EVA composite films with excellent electric/NIR driven reversible two-way shape memory properties.

Small and medium-sized enterprises (SMEs) are critical for achieving a green economy. This study aims to empirically explore business associations’ impacts on SMEs’ green transformation in China.

This study uses a multiple-case study approach. The authors collect data from four Chinese business associations and their members.

This study finds that business associations promote SMEs’ green transformation by providing individual and collective services. SMEs conduct green transformation by developing a green mindset and adopting green operations. The results show that individual services directly enhance members’ green mindset and green operations. Collective services promote members’ green mindset and green operations both directly and indirectly by building relational capital.

This study contributes to the literature by revealing that business associations play critical roles in assisting SMEs’ green transformation. In addition, this study suggests that SMEs may adopt different practices compared with large companies. The findings enhance the current understanding of how SMEs conduct green transformation and how business associations assist SMEs.

Continuous fiber-reinforced thermoplastic composites are a class of materials highly valuable for structural applications and modeling of heat transfer within them is critical to the design of their processing methods. However, the fiber reinforcement leads to highly anisotropic thermal conduction. Among a variety of methods to account for anisotropic thermal conductivity, continuum models with effective media approximation thermal conductivity are computationally efficient and require minimal data to begin modeling a specific composite material. The purpose of this study is to evalute the utility of these models.

In this work, six potential effective media approximation models are evaluated against experimental heating data. Thick (>25 mm) glass fiber-reinforced polyethylene terephthalate glycol (PET-G) specimens with 40% fiber volume fraction were heated with embedded resistance heating to produce validation and testing data sets. A two-dimensional finite-difference solver was implemented using each of the six effective media approximation models. The accuracy of each model is compared.

The model developed by Cheng and Vachon was found to predict the experimental results most accurately. Fit statistics were similar in the testing and validation data sets. This model is recommended for simulation of transient heating in continuous fiber-reinforced thermoplastic composites with low-to-moderate fiber volume fractions.

There are a wide variety of mathematical models for effective media approximation thermal conductivity, though very few have been applied to continuous fiber-reinforced thermoplastic composites. This work shows that the simplest methods based on rules of mixtures are well outperformed by more modern and complex models, and should be incorporated for accurate prediction of heating during thermal processing of fiber-reinforced thermoplastic composites.

This paper aims to develop an optimization model to enhance pipeline assembly performance. It focuses on predicting the pipeline’s assembly pose while ensuring compliance with clamp constraints.

The assembly pose of the pipeline is quantitatively assessed by a proposed indicator based on joint defects. The assembly interference between the pipeline and assembly boundary is characterized quantitatively. Subsequently, an analytical mapping relationship is established between the assembly pose and assembly interference. A digital fitting model, along with a novel indicator, is established to discern the fit between the pipeline and clamp. Using the proposed indicators as the optimization objective and penalty term, an optimization model is established to predict the assembly pose based on the reinforced particle swarm optimization, incorporating a proposed adaptive inertia weight.

The optimization model demonstrates robust search capability and rapid convergence, effectively minimizing joint defects while adhering to clamp constraints. This leads to enhanced pipeline assembly efficiency and the achievement of a one-time assembly process.

The offset of the assembly boundary and imperfections in pipeline manufacturing may lead to joint defects during pipeline assembly, as well as failure in the fit between the pipeline and clamp. The assembly pose predicted by the proposed optimization model can effectively reduce the joint defects and satisfy clamp constraints. The efficiency of pipeline modification and assembly has been significantly enhanced.

Cloud manufacturing platform has a high degree of openness, with a large variety of users having different needs. Designers on such platforms exhibit great differences in their knowledge abilities and knowledge needs, necessitating the cloud platform to provide personalized knowledge recommendation. To satisfy the personalized knowledge needs of the designers in product design tasks and other manufacturing tasks on a cloud manufacturing platform and provide them with high-quality knowledge resources, a knowledge recommendation method based on designers’ knowledge ability is proposed. The proposed method, with appropriate adjustments, can also be used for personalized knowledge recommendation to other personnel or institutions in cloud manufacturing platforms.

A knowledge recommendation method model is developed. The method consists of three stages. First, a designer knowledge system is constructed based on customer reviews in historical tasks, and designer knowledge ability and knowledge demand degree are quantitatively evaluated by synthesizing customer reviews and expert evaluations. Subsequently, the design knowledge domain ontology is constructed, and knowledge resources and tasks are modeled based on the ontology. Finally, the semantic similarity between tasks and knowledge resources and the knowledge demand degree of designers are integrated to calculate the knowledge recommendation coefficient, which realizes the personalized knowledge recommendation of designers.

Two design tasks of a 3D printing cloud platform are taken as examples to verify the feasibility and effectiveness of the proposed method. Compared with other methods, it is proved that the method proposed in this paper can obtain more knowledge resources that meet the needs of designers and tasks.

The method proposed in this paper is important for the expansion of data applications of the cloud manufacturing platform and for enriching the knowledge recommendation method. The proposed method has two innovations. First, both designer needs and task needs are considered in knowledge recommendation. Compared with most of the existing methods, which only consider one factor, this method is more comprehensive. Second, the designer’s knowledge ability model is constructed by using customer reviews on the cloud manufacturing platform. This overcomes the defect of low accuracy of the interest model in existing methods and makes full use of the big data of the cloud manufacturing platform.

This study aims to optimize the construction site layout planning (CSLP) problem, with a focus on prefabricated projects. It proposes the use of the oMOAHA algorithm, an enhanced version of the multi-objective artificial hummingbird algorithm (MOAHA), to address challenges related to search space exploration and local optimization in CSLP.

The study integrates three techniques – opposition-based learning (OBL), quasi-opposition and quasi-reflection – into the initialization phase of the MOAHA algorithm, creating the oMOAHA variant. This model is applied to all three types of CSLP problems – pre-determined location, grid system and continuous space – to evaluate its effectiveness. Six objective functions (three related to cost, two to safety and one to tower crane efficiency) and four site-related constraints are considered through three case studies taken from previous research and one real project involving prefabricated steel structures.

The oMOAHA algorithm demonstrates superior performance compared to previous models, consistently outperforming traditional approaches in CSLP optimization for prefabricated projects. In the real case study, the proposed model exceeded the actual project plan by 28–43%, indicating its potential to significantly improve both solution quality and project outcomes.

This study is the first to apply an optimization model to all three types of CSLP problems – pre-determined location, grid system and continuous space – within a unified framework. The integration of advanced techniques into the MOAHA algorithm and the model’s successful application in a real prefabricated project underscore its high applicability and effectiveness in modern construction management.