The quality and reliability in additive manufacturing is an emerging area. To ensure process quality and reliability, the influence of all process parameters and conditions needs to be understood. The product quality and reliability characteristics, i.e., dimensional accuracy, precision, repeatability, and reproducibility are mostly affected by inherent and systematic manufacturing process variations. This paper presents research on dimensional quality and distortion analysis of AlSi10Mg thin-walled parts developed by a selective laser melting technique. The input process parameters were fixed, and the impact of inherent process variation on dimensional accuracy and precision was studied. The process stability and variability were examined under repeatability and reproducibility conditions. The sample length (horizontal dimension) results revealed a 0.05 mm maximum dimensional error, 0.0197 mm repeatability, and 0.0169 mm reproducibility. Similarly, in sample height (vertical dimension) results, 0.258 mm maximum dimensional error, 0.0237 mm repeatability, and 0.0863 mm reproducibility were observed. The effect of varying design thickness on thickness accuracy was analyzed, and regression analysis performed. The maximum 0.038 mm error and 0.018 mm standard deviation was observed for the 1 mm thickness sample, which significantly decreased for sample thickness ≥2 mm. The % error decreased exponentially with increasing sample thickness. The distortion analysis was performed to explore the effect of sample thickness on part distortion. The 0.5 mm thickness sample shows a very high distortion comparatively, and it is reduced significantly for >0.5 mm thickness samples. The study is further extended to examine the effect of solution heat treatment and artificial aging on the accuracy, precision, and distortion; however, it did not improve the results. Conclusively, the sample dimensions, i.e., length and height, have shown fluctuations due to inherent process characteristics under repeatability and reproducibility conditions. The ANOVA results revealed that sample length means are not statistically significantly different, whereas sample height means are significantly different. The horizontal dimensions in the xy-plane have better accuracy and precision compared to the vertical dimension in the z-axis. The accuracy and precision increased, whereas part distortion decreased with increasing thickness.

The purpose of this study is to make good use of the massive amount of online user comment data to explore and analyze the dimensions of customer-perceived value and the importance of each dimension, given the background of China’s huge e-commerce market. We compiled a web crawler program to collect online comment data from online reviews. The crawled data were pre-processed and content analysis were performed. A customer-perceived value dictionary was constructed based on the extraction of frequent terms, literature review, and expert opinions. We re-identified the dimensions of customer-perceived value to include four key dimensions and corresponding subdivisions. Both the rationality and operability of the dimension model of customer-perceived value were validated and applied. Thereafter, the importance of various dimensions and the impacts of customer-perceived value dimensions on customer loyalty were analyzed and discussed. The empirical research results reveal that all four dimensions of customer-perceived value play an important role in customer-perceived value and that the patterns and degrees of the role of each dimension are rather different. Further, only certain parts of the dimensions of customer-perceived value have an impact on customer loyalty, and the degree of the impact differs substantially.

Product–service systems (PSSs) have great potential for competitiveness and sustainability. Customers’ requirements cannot be directly used in the design of a PSS. Accurate identification of customer requirements, especially hidden requirements in the product life cycle, and transformation of customer requirements into specific engineering characteristics for PSS design are urgent problems. This study proposed a systematic and whole-process framework employing specific identification processes and methods, as well as a big data analysis. A set of refined and integrated methods were used to better identify customer requirements and to transform the customer requirements into specific engineering characteristics more accurately and efficiently. We also used customers’ online review data—a huge information resource to be explored—and big data technology to improve the requirement information identification process. A case study was implemented to verify our methodology. We obtained the engineering characteristics of a smartphone PSS matching the customer requirements as well as the exact importance rankings of customer requirements and engineering characteristics. The analysis results revealed that the proposed methodology allowed PSS designers to assess the PSS requirements more specifically and accurately by providing an intuitive evaluation of the role and importance of the requirements, engineering characteristics, and their mutual interactions that were hidden or indirect.

Smart data selection can quickly sieve valuable information from initial data. Doing so improves the efficiency of analyzing situations to aid in better decision-making. Past methods have mostly been based on expert experience, which may be subjective and inefficient when dealing with large, complex datasets. Recently, the system analysis method has been exploited to find the key data. However, few studies address the indirect effects and heterogeneity of time series data. In this study, a data selection method, the modified Decision-Making Trial and Evaluation Laboratory (DEMATEL) method based on the objective data grey relational analysis (GRA), is used to enhance the ability to analyze time-series data. GRA was first applied to assess the direct impact in the raw data indicators. Then, a modified DEMATEL was adopted to find the overall impact by including the indirect impact and data heterogeneity. We applied the method to analyze the Commercial Modular Aero-Propulsion System Simulation (C-MAPSS) dataset and perform the remaining useful life (RUL) prediction of aircraft engines. The results suggest that our method predicts well. Our work offers a nuanced approach of identifying key information in time series data and has potential applications.

Sustainable development is of great significance. The emerging research on data-driven computational sustainability has become an effective way to solve this problem. This paper presents a fault diagnosis and prediction framework for complex systems based on multi-dimensional data and multi-method comparison, aimed at improving the reliability and sustainability of the system by selecting methods with relatively superior performance. This study took the avionics system in the industrial field as an example. Based on the literature research on typical fault modes and fault diagnosis requirements of avionics systems, three popular high-dimensional data-driven fault diagnosis methods—support vector machine, convolutional neural network, and long- and short-term memory neural network—were comprehensively analyzed and compared. Finally, the actual bearing failure data were used for programming in order to verify and compare various methods and the process of selecting the superior method driven by high-dimensional data was fully demonstrated. We attempt to provide a sustainable development idea that continuously explores multi-method integration and comparison, aimed at improving the calculation efficiency and accuracy of reliability assessments, optimizing system performance, and ultimately achieving the goal of long-term improvement of system reliability and sustainability.

In the process of Unmanned Aerial Vehicle (UAV) flight testing, plenty of compound faults exist, which could be composed of concurrent single faults or over-limit states alarmed by Built-In-Test (BIT) equipment. At present, there still lacks a suitable automatic labeling approach for UAV flight data, effectively utilizing the information of the BIT record. The performance of the originally employed flight data-driven fault diagnosis models based on machine learning needs to be improved as well. A compound fault labeling and diagnosis method based on actual flight data and the BIT record of the UAV during flight test phase is proposed, through labeling the flight data with compound fault modes corresponding to concurrent single faults recorded by the BIT system, and upgrading the original diagnosis model based on Gradient Boosting Decision Tree (GBDT) and Fully Convolutional Network (FCNN), to eXtreme Gradient Boosting (XGBoost), Light Gradient Boosting Machine (LightGBM) and modified Convolutional Neural Network (CNN). The experimental results based on actual test flight data show that the proposed method could effectively label the flight data and obtain a significant improvement in diagnostic performance, appearing to be practical in the UAV test flight process.