Forthcoming and Online First Articles

International Journal of Manufacturing Technology and Management

International Journal of Manufacturing Technology and Management (IJMTM)

Forthcoming articles have been peer-reviewed and accepted for publication but are pending final changes, are not yet published and may not appear here in their final order of publication until they are assigned to issues. Therefore, the content conforms to our standards but the presentation (e.g. typesetting and proof-reading) is not necessarily up to the Inderscience standard. Additionally, titles, authors, abstracts and keywords may change before publication. Articles will not be published until the final proofs are validated by their authors.

Forthcoming articles must be purchased for the purposes of research, teaching and private study only. These articles can be cited using the expression "in press". For example: Smith, J. (in press). Article Title. Journal Title.

Articles marked with this shopping trolley icon are available for purchase - click on the icon to send an email request to purchase.

Online First articles are published online here, before they appear in a journal issue. Online First articles are fully citeable, complete with a DOI. They can be cited, read, and downloaded. Online First articles are published as Open Access (OA) articles to make the latest research available as early as possible.

Open AccessArticles marked with this Open Access icon are Online First articles. They are freely available and openly accessible to all without any restriction except the ones stated in their respective CC licenses.

Register for our alerting service, which notifies you by email when new issues are published online.

We also offer which provide timely updates of tables of contents, newly published articles and calls for papers.

International Journal of Manufacturing Technology and Management (34 papers in press)

Regular Issues

  • Effect of Machining Parameters in Drilling of Glass Fiber Reinforced Polymer Composite with Modified AJM Process.   Order a copy of this article
    by Basanta Kumar Nanda, Debabrata Dhupal 
    Abstract: Conventional drilling of Glass Fiber Reinforced Polymer (GFRP) composite produces delamination effect due to peel-up and pull out of the layers. This research work represents drilling of GFRP rectangular workpieces with fluidized bed abrasive jet machining (FB-AJM) process using silicon carbide (SiC) abrasives. Experiments were conducted according to Box-Behnken design matrix to measure the four responses, namely, material removal rate (W), surface roughness (Ra), depth of cut (h), and taper angle (TA) by utilizing bed pressure (P), nozzle tip distance (Z) and abrasive grain size (G) as input parameters. Desirability based particle swarm optimization (PSO) was performed for the multi-objective optimization to find the optimal input parameters P, Z and G of 6 kgf/cm2, 7 mm, 260μm respectively. A single experiment was performed at the predicted optimal conditions to validate the model with minimum error. The SEM analysis of the machined composite surface revealed the initiation and propagation of cracks along with embedment of erodent particle.
    Keywords: Glass Fiber Reinforced Polymer (GFRP); fluidized bed abrasive jet machining (FB-AJM); silicon carbide (SiC); response surface methodology (RSM); Desirability-Function; particle swarm optimization (PSO).

  • FPGA Implementation of Gate Level Modified Adaptive Filter Architecture for Noise Cancellation Application   Order a copy of this article
    by Gomathi Swaminathan, Murugesan G 
    Abstract: This paper provides an approach for the implementation of modifications in the gate level in the adaptive filter architecture. An adaptive filter is appreciable when processing real time data and under the conditions when the signal statistics are not known a priori. Filtering data in real time needs a dedicated hardware. The resources available for implementing an efficient digital filtering algorithm are present in Field Programmable Gate Array. To design a dedicated hardware with low computational complexity, an approach of gate level modifications is incorporated in the EX-OR gate and the modified EX-OR gate is implemented in the coefficient decimation method. The proposed modified coefficient decimation method is applied in the adaptive algorithms such as Recursive Least Square, Affine Projection, and Kalman algorithm. The designed structures are experimented with the ECG and speech signal and implemented in the Xilinx Virtex 5 FPGA. Simulation and implementation results infer area reduction is achieved up to 50% and delay reduction of 10% is obtained in the proposed structures.
    Keywords: Field Programmable Gate Array; EX-OR; ECG; hardware; XILINX System Generator; full adder; half adder.
    DOI: 10.1504/IJMTM.2017.10011891
  • Study on Intelligent Control Method of Pharmaceutical Production Quality Based on Chromatographic Fingerprint   Order a copy of this article
    by Liyuan Zhang, Yong Yang, Xuefeng Wang 
    Abstract: Aiming at the problems of low control accuracy and long control time in traditional intelligent control methods of drug production quality, an intelligent control method of drug production quality based on chromatographic fingerprint was proposed. Extract the chromatographic fingerprint of drug production quality, and use wavelet transform to de-noise the obtained fingerprint. According to the preprocessing results, use the real-time mean method to calculate the mean value of the real-time converted chromatographic fingerprint data. Through the tolerance zone of the fingerprint, convert the drug chromatographic fingerprint data. Based on the data conversion results, grade the drug production quality through fuzzy classification, The interval hesitation fuzzy control chart is constructed to carry out intelligent control of drug production quality. The experimental results show that the proposed method has higher precision and shorter control time for drug production quality intelligent control.
    Keywords: Chromatographic fingerprint; Drug production quality; Intelligent control; Real-time mean method; Fuzzy control chart.
    DOI: 10.1504/IJMTM.2023.10056614
  • A method for surface wear detection of machined parts based on image processing   Order a copy of this article
    by Weilin Zeng, Weizhao Guo, Jiang Qiu, Hong Wen 
    Abstract: In order to reduce the error in the surface wear detection of machined parts, improve the detection accuracy of the failure degree of parts, and shorten the detection time, a method of surface wear detection of machined parts based on image processing is designed. Firstly, the wear mechanism is analyzed and the mathematical model of wear law is established; Then the relationship model of surface wear of machined parts is analyzed to determine the range of wear data to be detected. Finally, three components are used to determine the gray level image of part surface wear, calculate the pixel value of the maximum gray level image, and then binary processing is carried out. The noise is removed by means of the mean filtering algorithm. Finally, the wear detection of machined parts is realized by calculating the weighted mean. The results show that the wear detection error of the proposed method is low and has credibility.
    Keywords: image processing; Machined parts; Surface wear; Limit value; Mean filtering; Binarization.
    DOI: 10.1504/IJMTM.2023.10056615
  • Multi product supply chain scheduling method based on hybrid genetic algorithm   Order a copy of this article
    by Lingmin Yang 
    Abstract: In order to solve the shortcomings of traditional methods such as high Hamming loss value and low demand supply rate, a multi product supply chain scheduling method based on hybrid genetic algorithm was designed. Build supply time series, and sort the priority of supply chain scheduling according to the priority attribute values of backup plans in each link of the supply chain closed-loop model. Taking the shortest scheduling time as the objective function and the location constraints of the revolving warehouse as the constraints, the scheduling results are obtained by combining the genetic algorithm. In order to avoid the genetic algorithm falling into the local optimum, the simulated annealing algorithm is introduced to solve the global optimal solution of the objective function to achieve the coordinated scheduling of the supply chain. The experimental results show that this method can achieve multi product supply chain scheduling more reasonably and effectively.
    Keywords: Genetic algorithm; Simulated annealing algorithm; Multi product supply chain; Supply chain scheduling; Dispatch time; Positioning of turnover warehouse.
    DOI: 10.1504/IJMTM.2023.10056616
  • Research on the integrated logistics supply chain management model of foreign trade processing products   Order a copy of this article
    by Kui Wang, Fangfang Zhang, Jizhi Wang, Hongmei Zhao 
    Abstract: There are some problems in the integrated logistics supply chain management model of foreign trade processing products, such as high management risk coefficient, large model effectiveness evaluation error, and low supply chain evaluation accuracy. Therefore, this paper designs a method to evaluate the integrated logistics supply chain management mode of foreign trade processing products. Firstly, the characteristics of integrated logistics supply chain management mode for different foreign trade processing products are determined. Then, with the help of the analytic hierarchy process to determine the management model evaluation indicators and weights. Finally, determine the maximum demand of each node in the integrated logistics supply chain, build the model effectiveness evaluation model, and realise the evaluation of the supply chain management model. Experiments show that the proposed method can effectively reduce the risk coefficient and improve the accuracy of the model effectiveness evaluation.
    Keywords: foreign trade processing products; integrated supply chain; management mode; risk coefficient; analytic hierarchy process; judgement matrix.
    DOI: 10.1504/IJMTM.2023.10056617
  • Research on welding path modification of welding industrial robot based on leapfrog algorithm   Order a copy of this article
    by Fangwei Pan 
    Abstract: Aiming at the problems of large welding path correction error, low searching accuracy of industrial robot welding path, and long correction time, a welding path correction method based on leapfrog algorithm was proposed for welding industrial robot. Firstly, the key components of the industrial welding robot system and its interaction are determined, and its control system and operation principle are analyzed; Then calculate the total resistance and voltage change value of spot welding industrial robot, realize the change of position relationship in its running path by means of homogeneous transformation method, and determine the change of technical parameters of the robot under different rotation angles; Finally, the optimal path is searched globally, and the welding path of the welding industrial robot is modified by the idea of differential evolution. The results show that the proposed method can reduce the welding path correction error, and the correction time is short.
    Keywords: leapfrog algorithm; Welding industrial robot; Welding path; Correction method; Homogeneous transformation; global search.
    DOI: 10.1504/IJMTM.2023.10056618
  • Study on cost benefit estimation of enterprise environmental management based on multiple linear regression   Order a copy of this article
    by Mi Wang 
    Abstract: There are some problems in the cost benefit estimation of enterprise environmental management, such as poor precision of estimation results, low correlation of estimation indexes and long estimation time. Therefore, a cost benefit estimation method of enterprise environmental management based on multiple linear regression is proposed. First, the enterprise environmental management cost indicator system is constructed, and then the similarity between the indicator data is determined by using association rules and support calculation, and then combined and quantified. Finally, quantitatively describe the indicator data to determine that there is no correlation between them, remove the interference items in the indicator data, build an evaluation model with multiple linear regression, and verify the goodness of fit of the results through the sum of squares decomposition formula to complete the benefit estimation. The results show that the proposed method improves the accuracy of the estimation results, and the estimation speed is fast.
    Keywords: multiple linear regression; Enterprise environment; administration cost; Benefit estimation; Similarity; Quantitative description; Sum of squares decomposition.
    DOI: 10.1504/IJMTM.2023.10056619
  • Service quality evaluation of agricultural cold chain logistics supply chain based on k-means clustering algorithm   Order a copy of this article
    by YuXi Zhang  
    Abstract: In order to improve the reliability of the cold chain logistics supply chain and shorten the response time of the supply chain, a quality evaluation method of agricultural cold chain logistics supply chain based on k-means clustering algorithm was proposed. First of all, build the quality evaluation system of agricultural cold chain logistics supply chain. Secondly, the root method is selected to check the consistency of the judgment matrix, and the weight vector of the logistics supply chain quality evaluation is calculated. Finally, the k-means clustering algorithm is used to evaluate the supply chain service quality. The experimental results show that the supply chain quality reliability of this method is 0.98, and the service response time is only 6 minutes; The service satisfaction rate can reach 99.6%.
    Keywords: K-means clustering algorithm; Weight vector; Analytic Hierarchy Process; Agriculture products; Cold chain logistics; Service quality assessment.
    DOI: 10.1504/IJMTM.2023.10056620
  • Risk Assessment Method of Human Resources Outsourcing Based on Risk Matrix   Order a copy of this article
    by Yi Zhou  
    Abstract: In order to reduce the error of risk assessment and improve the correlation of risk indicators, a risk assessment method of human resource outsourcing based on risk matrix method is proposed. Determine the risk indicators of human resource outsourcing, use the rough set theory to determine the indiscernibility of the evaluation indicators, and use F-score to calculate the risk evaluation indicators and their information gain; Build a risk matrix coordinate chart to identify outsourcing risk factors, calculate the probability of outsourcing risk occurrence using the conditional random field joint probability formula, and build a risk assessment model through the risk matrix method to complete the risk assessment. The experimental results show that the evaluation error of the proposed method is always less than 2%, and the maximum correlation coefficient is close to 1, which verifies the effectiveness of the method.
    Keywords: Risk matrix method; Human resources; Outsourcing risk; F-score; Information gain; Joint probability of conditional random fields.
    DOI: 10.1504/IJMTM.2023.10056621
  • Real time location method of electric power material storage based on RFID technology   Order a copy of this article
    by Fei Li, Tuo Xin 
    Abstract: In order to overcome the problems of traditional warehousing location methods, such as time-consuming and low positioning accuracy, a real-time positioning method for electric power materials warehousing based on RFID technology is proposed. Firstly, the weight centroid between the electrical materials to be measured is determined by the trilateral measurement method to complete the distance measurement of electrical materials in storage. Secondly, determine the proportion of the weight of the location of electrical materials in storage, and set the electrical materials storage label with the help of Gaussian filter. Finally, RFID technology is introduced to analyze the probability of collision of electronic labels of warehousing materials. With the help of time stamp calculation, the collision of electronic labels is avoided and the warehousing location of materials is completed. The experimental results show that the positioning accuracy of this method is high, and the highest positioning accuracy reaches 92%.
    Keywords: RFID technology; Power material storage; Real time positioning; Gaussian filter; Electronic label.
    DOI: 10.1504/IJMTM.2023.10056623
  • Digital Packaging Design Method of Intelligent Products Based on Internet of Things Technology   Order a copy of this article
    by Gai Song  
    Abstract: Aiming at the problems of large feature extraction error and poor packaging image design effect in the process of digital packaging design of intelligent products at present, this paper designs a digital packaging design method for intelligent products based on the Internet of Things technology. First of all, build a digital packaging data collection platform for intelligent products based on the Internet of Things technology; Then, the digital packaging features of intelligent products are extracted by Harris corner feature method; Finally, the automatic generation model of packaging is constructed, and the automatic generation model is solved by maximizing the information entropy to complete the digital packaging design of intelligent products. The experimental results show that the minimum feature extraction error of the proposed method is 0.2%, and the maximum packaging color enhancement is 95%, which can effectively reduce the extraction error, and the digital packaging generation effect is good.
    Keywords: Internet of Things technology; Intelligent products; Digital packaging; Move window; Harris Corner Feature.
    DOI: 10.1504/IJMTM.2023.10056628
  • Design of real-time monitoring method for production line equipment status based on cloud computing and Internet of Things technology   Order a copy of this article
    by Pei Li, Jing Yuan, Yawen Hong, Yashe Lei 
    Abstract: In order to solve the problems of low monitoring accuracy and long monitoring time in the traditional real-time monitoring method of production line equipment status, this paper designs the real-time monitoring method of production line equipment status based on cloud computing and internet of things technology. Based on the perception of the internet of things labels, collect production line equipment status data and real-time upload, extract time domain parameters of production line equipment status, build historical memory matrix, establish and train multiple state estimation model, design distributed processing cloud computing platform MapReduce framework, complete real-time monitoring of production line equipment status under this framework. The experimental results show that the real-time monitoring accuracy of production line equipment status by the proposed method is 98%, the monitoring time is within 7.25 s, and it has the application effect of high precision and low time consumption.
    Keywords: cloud computing; internet of things technology; production line equipment; condition monitoring; radio frequency identification; RFID.
    DOI: 10.1504/IJMTM.2023.10059822
  • Analysis of the factors influencing the sustainable development of manufacturing industry under the wave of industry 4.0   Order a copy of this article
    by Xiaoying Bai, Yulong Wan 
    Abstract: As a pillar industry of the national economy, whether manufacturing industry can adapt to the environment of Industry 4.0 era and get sustainable development in this wave is a key issue in the field of manufacturing industry. Through the analysis of the background of Industry 4.0 wave, the research on the development status and problems of manufacturing industry, and the analysis of the influencing factors of the sustainable development of manufacturing industry, this paper makes clear the impact of environmental factors, investment factors and resource factors on the sustainable development of manufacturing industry under the Industry 4.0 wave. On the basis of the research results, we should formulate reasonable countermeasures for the sustainable development of manufacturing industry to promote the development of manufacturing industry in the new era.
    Keywords: Industry 4.0; manufacturing industry; sustainable development; environmental protection.
    DOI: 10.1504/IJMTM.2023.10059824
  • Transmission error compensation method for small module gears of CNC machine tools based on discretization decoupling calculation   Order a copy of this article
    by Huiqin Li, Jing Chang, Baofeng Zhang 
    Abstract: In order to improve the stability of gear transmission process and reduce transmission failure rate, a transmission error compensation method based on discretization decoupling calculation was proposed for small module gears of CNC machine tools. Analyze the dynamic changes in the gear transmission process, and set up sensors to collect the gear meshing frequency and transmission meshing frequency. Based on the sensing signal, establish a transmission error analysis model. The discretization analysis concept is introduced into the model, and the difference between the theoretical tooth surface and the actual machined tooth surface is taken as the transmission error. After the decoupling calculation, the error compensation amount is obtained.Experiment show that this method can control the gear transmission failure rate between 0.042 and 0.065, and maintain the transmission output power between 2800kW and 3000kW, indicating that this method can effectively achieve compensation for transmission errors.
    Keywords: CNC machine tools; Small module gear; Dynamics; Sensors; Transmission error; Discretization principle; Decoupling calculation; Error compensation.
    DOI: 10.1504/IJMTM.2023.10061700
  • Automatic Fault Diagnosis Method for Mining Hydraulic Support Based on Fuzzy Analysis   Order a copy of this article
    by Minjie Chen, Rongjin Yang 
    Abstract: A fuzzy analysis based automatic fault diagnosis method for mining hydraulic supports is proposed with the goal of improving fault detection rate, reducing misdiagnosis rate, and improving diagnostic efficiency. Firstly, for the structure of mining hydraulic support, real-time data collection of mining hydraulic support work is carried out through multiple DS18B20 sensor arrays; then, the moving average method is used to denoise the collected data, and the artificial bee colony algorithm is used to extract fault features; finally, based on the results of data denoising and feature extraction, a fuzzy rule library is established using expert experience and knowledge, and fault diagnosis is achieved through fuzzy reasoning and deblurring. The experimental results show that the highest fault detection rate of the proposed method is 83.5%, and the misdiagnosis rate remains below 2%, with a relatively short fault diagnosis time.
    Keywords: fuzzy analysis; hydraulic support; fault diagnosis; moving average method; fuzzy rule library; artificial bee colony.
    DOI: 10.1504/IJMTM.2023.10062073
  • Fault Information Identification Method of Industrial Production Equipment Based on Industrial internet of things   Order a copy of this article
    by Yun Yang  
    Abstract: In order to solve the problems of low recognition accuracy and long recognition time existing in the existing fault information recognition methods for industrial production equipment, this paper proposes a fault information recognition method for industrial production equipment based on the industrial internet of things. First, based on the industrial internet of things technology, build an information collection platform for industrial production equipment; then, based on wavelet coefficients, the equipment signal is pre-processed and the fault features of industrial production equipment are extracted based on sparse expression; finally, a least squares support vector model is constructed to classify fault signals and achieve recognition of industrial production equipment fault information. Through experiments, it can be seen that the accuracy of using the method proposed in this article for recognition is always above 96%, and the recognition time is always within 7.50 s, which has good recognition effect and efficiency.
    Keywords: industrial internet of things; IIoT; industrial production equipment; fault information; wavelet coefficients; sparse encoding; least squares support vector machine.
    DOI: 10.1504/IJMTM.2023.10062074
  • Optimization Method for Human Resource Scheduling in Manufacturing Industry Based on Decision Tree Algorithm   Order a copy of this article
    by Huan Wang, Hao Wang 
    Abstract: To overcome the problems of low efficiency in traditional labour resource allocation and poor measurement of employee competence, this paper designs a manufacturing human resource scheduling optimisation method based on decision tree algorithm. Firstly, determine the principles of resource scheduling optimisation and obtain scheduling optimisation parameters; then, using entropy calculation method to calculate the importance of parameters, establish a decision tree to obtain a classification scheduling optimisation rule set, and sort the parameter attributes; finally, the Gini index is used to classify the sample parameters and construct a decision tree for resource scheduling optimisation, achieving the final scheduling optimisation. The results show that the overall allocation efficiency under this method is higher than 98%, and the competency measurement level is all A, which improves the efficiency of labour resource allocation and has a good measurement of employee competency.
    Keywords: decision tree algorithm; manufacturing human resources; scheduling optimisation; employee competency; degree of collaboration; Gini index.
    DOI: 10.1504/IJMTM.2023.10062075
  • Research on automatic planning algorithm of surfacing repair path based on 3D vision technology   Order a copy of this article
    by Zhonghua Ni, Xinhua Wang 
    Abstract: To solve the problems existing in the traditional automatic planning algorithm of surfacing repair path, an automatic planning algorithm of surfacing repair path based on 3D vision technology is proposed. Firstly, the one-way planning distance is calculated, the pixel value containing weld in the target image is set to 0 based on 3D vision technology, the threshold value of weld is calculated, the surfacing repair path planning target is constructed, and the planning nodes are set. Calculate the displacement value of the component that needs welding deformation, and get the automatic planning result of surfacing repair path. The experimental results show that the target location result obtained by the proposed algorithm is more accurate, and the characteristics of the weld test sample image can be accurately planned. The average root mean square error after the supplement is smaller, which is 0.1 cm, indicating that the proposed method can effectively improve the path planning effect.
    Keywords: 3D vision technology; surfacing; repair the path; automatic planning algorithm; planning distance.
    DOI: 10.1504/IJMTM.2023.10062079
  • A Control Method for Disordered Sorting of Industrial Robots Based on Binocular Stereo Vision   Order a copy of this article
    by Shudong He 
    Abstract: Controlling the disorderly sorting of industrial robots can significantly improve sorting efficiency, reduce costs, and improve production efficiency. Therefore, this article proposes a method for disorderly sorting control of industrial robots based on binocular stereo vision. Firstly, the binocular stereo vision method is used to calibrate the sorting items of industrial robots; secondly, the binocular parallax principle is used to obtain the depth information of the selected items and extract the pose of the sorting target object; then, the rotation method is used to extract the minimum bounding rectangle, and the motion equation of the object to be sorted and the end of the robot is constructed; finally, implement unordered sorting control based on machine vision. The results show that the calibration error of the proposed method is only 1.6%, and the control accuracy is as high as 96.2%, indicating that the proposed method can effectively improve the sorting effect.
    Keywords: binocular stereo vision; unordered sorting; improve the A* algorithm; binocular parallax; depth information.
    DOI: 10.1504/IJMTM.2023.10062083
  • Scheduling method for industrial production processes in cloud manufacturing environment   Order a copy of this article
    by Yang Zhang, Meng-jun Huang, Chao Xu 
    Abstract: A method for scheduling industrial production processes in a cloud manufacturing environment is proposed with the aim of improving the scheduling effectiveness of industrial production processes. Establish a cloud manufacturing platform architecture, comprehensively consider industrial production tasks, processes, loading and unloading, and transportation links, and establish a scheduling framework in the cloud manufacturing environment; minimise industrial production time, production costs, and transportation scheduling frequency as the scheduling objective function, set production process constraints and production equipment constraints. Optimise the attraction term and random term of the standard firefly algorithm, and continuously optimise the objective function through multiple iterations to obtain the optimal scheduling result. The experimental results indicate that the production cost, transportation scheduling frequency, and industrial production time of this method are relatively low, indicating its feasibility.
    Keywords: cloud manufacturing; industrial production; production process scheduling; objective function; constraints; firefly algorithm.
    DOI: 10.1504/IJMTM.2023.10062084
  • Online recognition method for appearance defects in mechanical parts processing based on machine vision   Order a copy of this article
    by Qian Meng, Pengfei Shi 
    Abstract: In order to solve the shortcomings of traditional parts appearance defect recognition methods with low recognition accuracy and long recognition time, this paper proposes an online recognition method for mechanical parts processing appearance defects based on machine vision. Firstly, the image acquisition environment of mechanical parts based on machine vision is determined. Secondly, the part image is pre-processed; thirdly, the maximum entropy segmentation method is used to complete the image segmentation. Finally, the defect texture features of the part image are extracted, and the support vector machine algorithm is combined to realise the online recognition of the appearance defects of mechanical parts. Experiments show that the recognition time of the proposed method never exceeds 600 s, the recognition accuracy is 93.75%, and the average time overhead of identifying a part is 0.5 s, which has high recognition accuracy and less time overhead, and has better application performance.
    Keywords: machine vision; mechanical parts; cosmetic imperfections; online recognition; variable threshold technology; maximum entropy segmentation.
    DOI: 10.1504/IJMTM.2023.10062086
  • Product Design Sketch Defect Detection Method Based on Feature Extraction   Order a copy of this article
    by Liwei Sun 
    Abstract: There are problems with high false alarm rates and low registration of defect features in product design sketch defect detection. To this end, a defect detection method for product design sketches based on feature extraction is designed. First, the Gaussian filter method is introduced to remove the noise in the product design sketch, and grayscale processing is carried out. Secondly, linear transformation of product design sketches expands the grayscale range of sketch defects and changes the local enhancement effect. Finally, through the PCA algorithm in the feature extraction algorithm, the defect features of the product design sketch are extracted, and the product design sketch with this feature is regarded as an image with defects. The results show that the proposed method can effectively reduce the false alarm rate in detection and improve the registration degree of defect features, with a maximum registration degree of about 99%.
    Keywords: feature extraction; product design sketch; defect detection; grayscale; PCA algorithm; defect feature extraction.
    DOI: 10.1504/IJMTM.2023.10062087
  • A Method for Predicting the Remaining Life of Mechanical Equipment in Production Lines Based on Similarity Features   Order a copy of this article
    by Xiangyang Mei, Tianbiao Yang, Wenyou Gao 
    Abstract: In order to shorten the time required for predicting the remaining life of mechanical equipment and reduce prediction errors, this paper proposes a method for predicting the remaining life of mechanical equipment in production lines based on similarity features. Firstly, obtain the vibration signals of mechanical equipment in the production line and extract signal features; then, calculate the similarity characteristics between the degradation indicators of mechanical equipment. Finally, the DTW method is used to measure the similarity of the overall lifespan of mechanical equipment, and the final remaining lifespan of the predicted samples is calculated based on the actual remaining lifespan of each reference sample and corresponding weights, achieving residual lifespan prediction. The results show that the prediction time of our method is only four seconds, and the prediction error does not exceed 8.33%, which verifies the effectiveness of our method.
    Keywords: similarity feature; DTW method; normalisation processing; local similarity; EEMD.
    DOI: 10.1504/IJMTM.2023.10062088

Special Issue on: LISS16-OPMSC Optimisation Problems in Manufacturing and Supply Chains

    by Manivel Palanisamy 
    Abstract: Objective The aim of the study was to identify the pitfalls in the existing supply chain of private and hospital pharmacies, identify the variables that influencing more in the efficiency of the pharmacy, select the appropriate inventory analysis and suggest the prioritized inventory matrix to meet customer requirement in efficient manner. Methods The qualitative and quantitative data has been collected through questionnaire and interview of the pharmacy personnel. Through SPSS, the normal distribution and correlation analysis to be carried out on sample data. The pitfalls were solved by the selection of appropriate inventory analysis and suggesting the prioritized inventory matrix to narrowed down the drugs for monitoring and control strategies of pharmacy drugs. Results The normal distribution and correlation analysis results shows that the variables collected through questionnaire are normally distributed and the variables that most influence the efficiency of the pharmacy supply chain. The implementation of Prioritized ABC FSN matrix in pharmacy will helps to provide high-quality service to the customer and there will be an adequate supply of the items in the pharmacies. The future work also to be discussed. Conclusion The prioritized ABC FSN matrix to be adopted as a routine practice for optimal use of resources and elimination of out-of-stock and over stock situations in the hospital and private pharmacy.
    Keywords: Pharmacy Supply Chain; Inventory Management System; Stock Outs; Overstock; IT Support in Pharmacy; correlation analysis; Inventory analysis.

Special Issue on: Big Data and AI for Process Innovation in the Industry 4.0 Era

  • Research on total amount prediction of import and export trade data based on combined predicting model   Order a copy of this article
    by Jianxin Chen, Xiaoke Zhao 
    Abstract: In order to overcome the problems of low index significance, low prediction efficiency and low prediction accuracy in the process of prediction of import and export trade data, a new research method for total amount prediction of import and export trade data based on combined prediction model. Combine neural network prediction method and support vector machine prediction method to build a combined prediction model. Through the back propagation process, it can reduce the error between the expected and the actual output, and modify the network threshold and weight, so as to realize the classification of samples in high-dimensional space in support vector machine, and the two methods are combined to complete the prediction of total import and export trade data. The experimental results show that the significance of the proposed method is close to 100%, the prediction time is within 0.5 min, and the prediction accuracy is more than 90%.
    Keywords: Combination predicting model; Import and export trade; Data predicting; Influencing factors.

Special Issue on: Internet of Things and Big Data on Industrial Production and Operation Applications

  • Economic investment risk prediction model and algorithm based on data mining method   Order a copy of this article
    by Yi Chen 
    Abstract: Investment decisions have a broad and far-reaching impact on the operating conditions of the entire enterprise. Once the investment decision is wrong, it will bring huge risks. As a data analysis technology, data mining can simulate mathematical models or algorithms by analysing historical data, which greatly improves the accuracy of prediction. The purpose of this paper is to study the application of data mining technology in the field of investment management. This paper constructs an economic investment risk prediction model based on data mining. The research results show that the sensitive factor affecting the investment status in the model obtained by the data mining algorithm is the quick ratio. When the quick ratio is less than or equal to 1.603, the investment is one year; if the result is greater than 1.603, the investment is five years.
    Keywords: investment risk; data mining; cluster analysis; factor analysis; decision tree.
    DOI: 10.1504/IJMTM.2022.10051525
  • Image dehazing network based on improved convolutional neural network   Order a copy of this article
    by Changxiu Dai 
    Abstract: Image dehazing enhances its quality by restoring the actual pixels influenced by poor light and intensity due to environmental and other factors. Hazy images are rectified to improve visibility, guidance, and object recognition through channel attribute corrections. This article introduces a pre-emptive dehazing network (PDN) using an improved convolutional neural network (ICNN) for single to multi-image dehazing. In the proposed method, neural network layers are operated for intensity-based single and multi-feature analysis. The image is split based on intensity pixels for identifying the channel corrections. This channel correction and intensity verifications are processed using CNN in different independent layers. In the CNN training, the channel correction from the hidden layers and pixel correlation from the external dataset is performed for dehazing the image pixels. The dehazed pixels are organised based on the original input organisation for verifying the similarity measure. The proposed method's performance is validated utilising the metrics similarity, error, precision, F1-score, and time complexity.
    Keywords: channel correction; convolutional neural network; CNN; image dehazing; pixel correlation; pre-emptive dehazing network; PDN.
    DOI: 10.1504/IJMTM.2022.10051436
  • Design of computer image automatic processing system based on artificial intelligence algorithm   Order a copy of this article
    by Guoqiang You 
    Abstract: Automatic image processing systems are applied for recognising human faces in crowds, person identification, and face matching applications. The varying textures, input representation, and position impact detection accuracy and recognition. Therefore, this article introduces an automatic image processing method (AIPM) for face recognition (FR) using deep learning (DL) paradigm. This method extracts the textural features based on the image position and classifies them based on pixel mapping. Semantic (even) and uneven pixel variations are observed in the classification process. The semantic classified pixels are used for correlating different image segments that are further used for training the learning network. The uneven pixels classified using DL is discarded to prevent recognition errors. The DL paradigm verifies the pixel position and coordinate mapping between different inputs. The detection is improved based on the classified output for semantic and uneven pixels. The training is based on semantic and mapping pixels, for which the training is improvised using erroneous pixels. Therefore, precision is improved with controlled analysis complexity.
    Keywords: artificial intelligence; automatic processing; face recognition; feature extraction; automatic image processing method; AIPM.
    DOI: 10.1504/IJMTM.2022.10052039
  • Fuzzy system for image defect detection based on machine vision   Order a copy of this article
    by Yiqiang Lai, Yongjun Qi, Xianfeng Zeng 
    Abstract: With the continuous upgrading of the industrial field, the market's requirements for product quality are also increasing, which requires more accurate product monitoring equipment. This paper analyses the composition of the vision system, and compares the collected images with the defects under manual detection based on the non-local mean denoising algorithm, and the results meet the system requirements. The experimental results show that the image size is 140*141, the distortion rate is 0.992, the image size is 120*81, and the distortion rate is 0.703. This means that the larger the image size, the higher the distortion. Before the improved algorithm, a total of 47 defects were detected, while after the improved algorithm, a total of 83 defects were detected. It can be seen that when the algorithm is improved, the number of defect detections increases significantly.
    Keywords: machine vision; image processing; image defect detection; fuzzy systems.
    DOI: 10.1504/IJMTM.2022.10051413
  • Pre-settlement audit in project cost management based on cloud computing   Order a copy of this article
    by Daibing Cheng, Zhongyan Luo, Rong Ju, Yong Zhang 
    Abstract: This research mainly analysed the audit method of pre-settlement in project cost management, which served as a reference for the work in this field. The basis of the pre-settlement audit was the specific information of construction material submitted by the construction party to ensure its authenticity and legitimacy, so that the benefits of investment could be fully exerted, and the ultimate goal of resources optimal allocation could be achieved. Because the pre-settlement audit was an extremely important process in the standardised project cost, there was a need to improve the audit level so that the effective management objectives could be achieved. Therefore, we introduced cloud computing to provide a certain basis for reducing audit cost effectively. The experimental results showed that the application of the cloud computing to the pre-settlement audit had produced positive effects, and the accuracy of the audit calculation could reach about 91%.
    Keywords: cloud computing; project cost management; pre-settlement work; review method.
    DOI: 10.1504/IJMTM.2022.10060191
  • Automatic human face recognition system of image processing based on BP neural network paradigm   Order a copy of this article
    by Pei Yang, Guoqiang You 
    Abstract: Observation video analysis is useful in recognising human faces in crowded and coinciding scenarios. Overlapping images result in false recognition due to non-semantic textural features. The boundary analysis varies for this process, generating segments exceeding masks of the original image. Backpropagation learning (BPL) based textural-edge detection and recognition model (TED-RM) is designed to resolve this issue. The proposed model exploits the masked and un-masked textural features for identifying the semantics of the input. After this identification process, appropriate features are analysed for semantics and correlation with the inward and overlapping video image input edges. The masked and un-masked regions' semantic features are recurrently correlated with the previous datasets for independent human faces. The mapping feature points are identified and correlated with the actual edge of the training input. The non-semantic edge points are classified for further training and validation to detect errors in further input analysis. The proposed TED-RM improves 10.84% high accuracy, 11.5% less processing time, 10.2% high true positives, 5.55% less error, and 10.6% high recall compared to existing methods.
    Keywords: DRL; face recognition; image semantics; texture classification; video analytics.
    DOI: 10.1504/IJMTM.2022.10055027
  • Simulation of EPC consortium partnership stability and data based on prospect theory   Order a copy of this article
    by Judan Hu, Shuang Tang, Minjie Yang 
    Abstract: For the rapid promotion and healthy development of the EPC general contracting mode, this paper constructs a game model for the stability evolution of the EPC consortium tripartite cooperation relationship based on the prospect theory from the perspective of the EPC consortium led by the design unit, analyses the behavioural interaction mechanism of the game subjects and the factors affecting the system evolution, and uses system dynamics for data simulation. The results show that the rate at which the EPC consortium triad eventually evolves into a stable cooperation strategy is positively related to the initial probability, loss avoidance coefficient, gain sensitivity, and gain/loss sharing coefficient, and negatively related to the risk pursuit coefficient. The steady state of the system is influenced by the strength of penalty and the level of regulation/cooperation cost. This paper can provide insights and a theoretical basis for the long-term stable development of EPC consortium.
    Keywords: EPC consortium; cooperative stability; evolutionary games; prospect theory; system dynamics.
    DOI: 10.1504/IJMTM.2023.10056209
  • Investigation on security management risk assessment of accounting resource sharing under internet of things big data technology   Order a copy of this article
    by Jiandong Wang 
    Abstract: While accounting resource sharing is convenient and saves a lot of time and energy, it also brings many potential risks to the enterprise. Therefore, how to protect and manage accounting resource sharing has become a top priority. This paper mainly studies the possible risks in the accounting resource sharing life cycle and the assessment of the security management risk of accounting resource sharing under the big data technology of the internet of things. This article will use various methods to evaluate the risk of accounting resource sharing security management, use information entropy and analytic hierarchy process to evaluate the uncertainty and risk size of accounting resource sharing security management risk, and use Markov chain and Bayesian formula to calculate the risk probability. The results show that the security protection efficiency of the optimised accounting resource sharing system has increased by 6.13%, and security management risks have been effectively controlled.
    Keywords: internet of things; big data; resource sharing; risk assessment; information entropy; analytic hierarchy process; AHP; Markov chain; Bayes rule.
    DOI: 10.1504/IJMTM.2022.10060980