Search results

The purpose of this study was conducted at an enterprise that produces fasteners and is one of the leading companies in the sector in terms of market share. Possible defects in the coating of bolts and nuts either lead to products being scrapped or all of the coating process being repeated from beginning to end. In both cases, the enterprise faces a waste of time and excessive costs. Through this project, the six sigma theory and its means were effectively used to improve the efficiency and quality management of the company. The selection of the six sigma project has also contributed to the creation of various documents to be used for project screening and evaluation of financial results.

Six sigma is an optimization strategy that is used to improve the profitability of businesses, avoid waste, scrap and losses, reduce costs and improve the effectiveness of all activities to meet or exceed customers’ needs and expectations. Six sigma’s process improvement model, known as Definition-Measurement-Analysis-Improvement-Control, contributes to the economic and technical achievements of businesses. The normal distribution of a process should be within ±3 sigma of the mean. This represents a scale of 99.7% certainty. However, improving the process through the utilization of the six sigma rule, which accepts normal variabilities of processes twice as strict, will result in an error rate of 3.4 per million instead of 2,700 per million for each product or service.

Using six sigma practices to reduce the costs associated with low quality and to increase economic added value became a cultural practice. With this, the continuation of six sigma practices throughout the Company was intended. The annual cost reduction achieved with the utilization of six sigma practices can be up to $21,780. When time savings are also considered, a loss reduction of about $30,000 each year can be achieved. The coating thickness efficiency increased from 85% to 95% after the improvements made through the six sigma project. There is a significant increase in the efficiency of coating thickness. In addition, the coating thickness efficiency is also close to the target value of 95%–97%.

The results of the study were optimized with the help of deep learning. The performance of the model created in deep learning was quite close to the actual performance. This result implicates the validity of the improvement work. The results may act as a guide for the use of deep learning in new projects.

Fiber reinforced additive manufacturing (FRAM) is an emerging technology that combines additive manufacturing and composite materials. As a result, design freedom offered by the manufacturing process can be leveraged in design optimization. The purpose of the study is to propose a novel method that improves structural performance by optimizing 3D print orientation of FRAM components.

This work proposes a two-part design optimization method that optimizes 3D global print orientation and topology of a component to improve a structural objective function. The method considers two classes of design variables: (1) print orientation design variables and (2) density-based topology design variables. Print orientation design variables determine a unique 3D print orientation to influence anisotropic material properties. Topology optimization determines an optimal distribution of material within the optimized print orientation.

Two academic examples are used to demonstrate basic behavior of the method in tension and shear. Print orientation and sequential topology optimization improve structural compliance by 90% and 58%, respectively. An industry-level example, an aerospace component, is optimized. The proposed method is used to achieve an 11% and 15% reduction of structural compliance compared to alternative FRAM designs. In addition, compliance is reduced by 43% compared to an equal-mass aluminum design.

Current research surrounding FRAM focuses on the manufacturing process and neglects opportunities to leverage design freedom provided by FRAM. Previous FRAM optimization methods only optimize fiber orientation within a 2D plane and do not establish an optimized 3D print orientation, neglecting exploration of the entire orientation design space.

Adsorption parameters (e.g. Langmuir constant, mass transfer coefficient and Thomas rate constant) are involved in the design of aqueous-media adsorption treatment units. However, the classic approach to estimating such parameters is perceived to be imprecise. Herein, the essential features and performances of the ant colony, bee colony and elephant herd optimisation approaches are introduced to the experimental chemist and chemical engineer engaged in adsorption research for aqueous systems. Key research and development directions, believed to harness these algorithms for real-scale water treatment (which falls within the wide-ranging coverage of the Sustainable Development Goal 6 (SDG 6) ‘Clean Water and Sanitation for All’), are also proposed. The ant colony, bee colony and elephant herd optimisations have higher precision and accuracy, and are particularly efficient in finding the global optimum solution. It is hoped that the discussions can stimulate both the experimental chemist and chemical engineer to delineate the progress achieved so far and collaborate further to devise strategies for integrating these intelligent optimisations in the design and operation of real multicomponent multi-complexity adsorption systems for water purification.

This paper aims to propose a framework for optimizing the pose in the assembly process of the non-ideal parts considering the manufacturing deviations and contact deformations. Furthermore, the accuracy of the method would be verified by comparing it with the other conventional methods for calculating the optimal assembly pose.

First, the surface morphology of the parts with manufacturing deviations would be modeled to obtain the skin model shapes that can characterize the specific geometric features of the part. The model can provide the basis for the subsequent contact deformation analysis. Second, the simulated non-nominal components are discretized into point cloud data, and the spatial position of the feature points is corrected. Furthermore, the evaluation index to measure the assembly quality has been established, which integrates the contact deformations and the spatial relationship of the non-nominal parts’ key feature points. Third, the improved particle swarm optimization (PSO) algorithm combined with the finite element method is applied to the process of solving the optimal pose of the assembly, and further deformation calculations are conducted based on interference detection. Finally, the feasibility of the optimal pose prediction method is verified by a case.

The proposed method has been well suited to solve the problem of the assembly process for the non-ideal parts with complex geometric deviations. It can obtain the reasonable assembly optimal pose considering the constraints of the surface morphological features and contact deformations. This paper has verified the effectiveness of the method with an example of the shaft-hole assembly.

The method proposed in this paper has been well suited to the problem of the assembly process for the non-ideal parts with complex geometric deviations. It can obtain the reasonable assembly optimal pose considering the constraints of the surface morphological features and contact deformations. This paper has verified the method with an example of the shaft-hole assembly.

The different surface morphology influenced by manufacturing deviations will lead to the various contact behaviors of the mating surfaces. The assembly problem for the components with complex geometry is usually accompanied by deformation due to the loading during the contact process, which may further affect the accuracy of the assembly. Traditional approaches often use worst-case methods such as tolerance offsets to analyze and optimize the assembly pose. In this paper, it is able to characterize the specific parts in detail by introducing the skin model shapes represented with the point cloud data. The dynamic changes in the parts' contact during the fitting process are also considered. Using the PSO method that takes into account the contact deformations improve the accuracy by 60.7% over the original method that uses geometric alignment alone. Moreover, it can optimize the range control of the contact to the maximum extent to prevent excessive deformations.

Because of the key role of joint torque in industrial robots (IRs) motion performance control and energy consumption calculation and efficiency optimization, the purpose of this paper is to propose a deep learning torque prediction method based on long short-term memory (LSTM) recurrent neural networks optimized by particle swarm optimization (PSO), which can accurately predict the the joint torque.

The proposed model optimized the LSTM with PSO algorithm to accurately predict the IRs joint torque. The authors design an excitation trajectory for ABB 1600–10/145 experimental robot and collect its relative dynamic data. The LSTM model was trained with the experimental data, and PSO was used to find optimal number of LSTM nodes and learning rate, then a torque prediction model is established based on PSO-LSTM deep learning method. The novel model is used to predict the robot’s six joint torque and the root mean error squares of the predicted data together with least squares (LS) method were comparably studied.

The predicted joint torque value by PSO-LSTM deep learning approach is highly overlapped with those from real experiment robot, and the error is quite small. The average square error between the predicted joint torque data and experiment data is 2.31 N.m smaller than that with the LS method. The accuracy of the novel PSO-LSTM learning method for joint torque prediction of IR is proved.

PSO and LSTM model are deeply integrated for the first time to predict the joint torque of IR and the prediction accuracy is verified.

In order to solve the problems faced by First Order Reliability Method (FORM) and First Order Saddlepoint Approximation (FOSA) in structural reliability optimization, this paper aims to propose a new Reliability-based Design Optimization (RBDO) strategy for offshore engineering structures based on Original Probabilistic Model (OPM) decoupling strategy. The application of this innovative technique to other maritime structures has the potential to substantially improve their design process by optimizing cost and enhancing structural reliability.

In the strategy proposed by this paper, sequential optimization and reliability assessment method and surrogate model are used to improve the efficiency for solving RBDO. The strategy is applied to the analysis of two marine engineering structure cases of ship cargo hold structure and frame ring of underwater skirt pile gripper. The effectiveness of the method is proved by comparing the original design and the optimized results.

In this paper, the proposed new RBDO strategy is used to optimize the design of the ship cargo hold structure and the frame ring of the underwater skirt pile gripper. According to the results obtained, compared with the original design, the structure of optimization design has better reliability and stability, and reduces the risk of failure. This optimization can also better balance the relationship between performance and cost. Therefore, it is recommended for related RBDO problems in the field of marine engineering.

In view of the limitations of FORM and FOSA that may produce multiple MPPs for a single performance function, the new RBDO strategy proposed in this study provides valuable insights and robust methods for the optimization design of offshore engineering structures. It emphasizes the importance of combining advanced MPP search technology and integrating SORA and surrogate models to achieve more economical and reliable design.

This paper aims to present a multi-fidelity surrogate-based optimization (MFSBO) method for computationally accurate and efficient design of microfluidic concentration gradient generators (µCGGs).

Cokriging-based multi-fidelity surrogate model (MFSM) is constructed to combine data with varying fidelities and computational costs to accelerate the optimization process and improve design accuracy. An adaptive sampling approach based on parallel infill of multiple low-fidelity (LF) samples without notably adding computation burden is developed. The proposed optimization framework is compared with a surrogate-based optimization (SBO) method that relies on data from a single source, and a conventional multi-fidelity adaptive sampling and optimization method in terms of the convergence rate and design accuracy.

The results demonstrate that proposed MFSBO method allows faster convergence and better designs than SBO for all case studies with 49% more reduction in the objective function value on average. It is also found that parallel infill (MFSBO-4) with four LF samples, enables more robust, efficient and accurate designs than conventional multi-fidelity infill (MFSBO-1) that only adopts one LF sample during each iteration for more complex optimization problems.

A MFSM based on cokriging method is constructed to utilize data with varying fidelities, accuracies and computational costs for µCGG design. A parallel infill strategy based on multiple infill criteria is developed to accelerate the convergence and improve the design accuracy of optimization. The proposed methodology is proved to be a feasible method for µCGG design and its computational efficiency is verified.

This research intends to develop a model for predicting employees at a high-risk attrition and identify the most important factors affecting them.

In this study, using the triangulation technique of a mixed research method, the employee attrition problem is investigated by identifying its affecting factors. For that matter, data related to the human resources department of a pharmaceutical company in Iran are used. And to achieve the intended goal, advanced data mining algorithms and interviews with human resource managers are applied.

A model for predicting employees at a high-risk attrition is presented based on the gradient boosting machine algorithm with 89% accuracy. The use of the mixed research approach shows that qualitative and quantitative methods can be more effective in identifying the factors affecting employee churn or loss of staff. The results also contain a new situation arising out of the COVID-19 pandemic and remote working scenarios having impact on employee attrition. Finally, human resource policies are presented based on variables related to each of the identified factors.

The novel contributions of this study include real data related to a leading pharmaceutical company as well as a combination of two quantitative and qualitative methods. The hybrid approach can identify the reasons for attrition and, consequently, retention policies to benefit from the advantage of both approaches. Data mining can be useful to identify the factors, which are usually not mentioned in termination interviews, such as direct managers. On the other hand, the results obtained from termination interviews can also include features that the authors cannot identify through data mining, which are specifically related to the characteristics of the pharmaceutical industry such as building a more professional career path. From a practical perspective, since this company specializes in pharmaceutical marketing in a new way and is primarily comprised graduates, it is important to note that the churn of specialized people disperses organizational and technological know-how. On the other hand, the pharmacist community in Iran is small, and their attrition might adversely affect not only the reputation of an organization but the employer's brand as well. So, this research would help other similar firms in retaining their valuable human capital.

This paper aims to use convolutional neural networks (CNNs) to provide an objective approach to classify deteriorated building assets according to the type and extent of damage. This research supports automated inspection of buildings and focuses on roofing elements as one of the most critical and externally distressed elements in buildings.

In this paper, 5,000+ images of deteriorated roofs from several buildings were collected to design a CNN system that automatically identifies and sizes roofing defects. Experimenting with different CNN formulations, the best accuracy is achieved using two-stage CNNs. The first-stage CNN classifies images into defect/no defect, while the second stage classifies the defected images according to the damage type. Based on the image classification, optimization is used to prioritize roof repairs by maximizing the return from limited rehabilitation funds.

The developed CNNs reached 95% and 97% accuracy for the first and second phases, respectively, which is higher than achieved in previous literature efforts. Using the proposed model to automate inspection and condition assessment activities proved to be faster than conventional methods. Repair/replace strategy for a case study of 21 campus buildings based on their condition and budgetary constraints was suggested.

Future research includes testing different data acquisition technologies (e.g. infrared imaging), performing severity-based classification and integrating with BIM for defect localization.

This study provides an objective approach to automate asset condition assessment and improve funding decisions using a combination of image analysis and optimization techniques. The proposed approach is applicable toward other asset types and components.

Recent United Nations Climate Change Conferences recognise extreme climate change of heatwaves, floods and droughts as threatening risks to the resilience and success of public–private partnership (PPP) infrastructure projects. Such conferences together with available project reports and empirical studies recommend project managers and practitioners to adopt smart technologies and develop robust measures to tackle climate risk exposure. Comparatively, artificial intelligence (AI) risk management tools are better to mitigate climate risk, but it has been inadequately explored in the PPP sector. Thus, this study aims to explore the tools and roles of AI in climate risk management of PPP infrastructure projects.

Systematically, this study compiles and analyses 36 peer-reviewed journal articles sourced from Scopus, Web of Science, Google Scholar and PubMed.

The results demonstrate deep learning, building information modelling, robotic automations, remote sensors and fuzzy logic as major key AI-based risk models (tools) for PPP infrastructures. The roles of AI in climate risk management of PPPs include risk detection, analysis, controls and prediction.

For researchers, the findings provide relevant guide for further investigations into AI and climate risks within the PPP research domain.

This article highlights the AI tools in mitigating climate crisis in PPP infrastructure management.

This article provides strong arguments for the utilisation of AI in understanding and managing numerous challenges related to climate change in PPP infrastructure projects.