دانلود مقاله تجزیه و تحلیل پایداری زنجیره تامین با پیش‌بینی انتشار ذرات معلق
چکیده
1. مقدمه
2. مرور مطالعات پیشین
3. چارچوب تحلیلی یکپارچه برای تجزیه و تحلیل پایداری مدیریت زنجیره تامین
4. نتایج
5. نتیجه گیری
منابع
Abstract
1. Introduction
2. Literature review
3. Integrated analytical framework for sustainability analysis of supply chain management
4. Results
5. Conclusions
Disclosure statement
References
چکیده
با افزایش اهمیت انتشار آلاینده های هوا برای اقتصاد پلت فرم و مدیریت زنجیره تامین سبز، تجزیه و تحلیل روند و همبستگی بین انتشار ذرات معلق و آمار زنجیره تامین ضروری است. رویکردهای معمولی پیش‌بینی ذرات معلق را با تحلیل پایداری ادغام نمی‌کنند و از مسائل رایجی مانند دقت طبقه‌بندی پایین و عملکرد پیش‌بینی ناپایدار رنج می‌برند. در این مطالعه، ما یک چارچوب تحلیلی یکپارچه برای تحلیل پایداری مدیریت زنجیره تامین از طریق پیش‌بینی انتشار ذرات معلق پیشنهاد می‌کنیم. به طور خاص، ما روند عملکرد و تجزیه و تحلیل همبستگی بین انتشار ذرات معلق (PM2.5 و PM10) و آمار زنجیره تامین در پکن چین را انجام دادیم. ما الگوریتم تقویت و روش شبکه عصبی را برای پیش‌بینی انتشار ذرات ترکیب می‌کنیم. نتایج تجربی نشان می‌دهد که مدل پیش‌بینی ما عملکرد بالایی دارد. تجزیه و تحلیل پایداری نشان می دهد که رشد مداوم عملیات زنجیره تامین با کاهش انتشار آلاینده های هوا در چین همراه است.
توجه! این متن ترجمه ماشینی بوده و توسط مترجمین ای ترجمه، ترجمه نشده است.
Abstract
With the increasing importance of air pollutant emissions to the platform economy and green supply chain management, it is essential to analyse the trend and correlation between particulate matter emissions and supply chain statistics. Typical approaches do not integrate particulate matter prediction with the sustainability analysis, and suffer from common issues such as low classification accuracy and unstable prediction performance. In this study, we propose an integrated analytical framework for sustainability analysis of supply chain management through particulate matter emissions prediction. Specifically, we performance trend and correlation analysis between particulate matter emissions (PM2.5 and PM10) and supply chain statistics in Beijing of China. We combine the boosting algorithm and neural network method to predict particulate matter emissions. Experimental results show that our prediction model achieved high performance. Sustainability analysis shows that the steady growth of the supply chain operations is accompanied by decreasing air pollutant emissions in China.
Introduction
Today, human society is facing very severe environmental and resource problems. The green and sustainable supply chain are the concrete manifestations of sustainable development strategies that comprehensively consider these two issues in platform economy management. In recent years, continued urbanisation and industrialisation, particularly in developing countries, has led to severe deterioration in air quality and a rapid increase in the degree of contamination (Shi and Yu 2020). Air pollution has substantially affected the living environment of the human population and endangered health (Stieb et al. 2019). The United Nations (UN) issued Sustainable Development Goals (SDGs) in 2015, serving as a blueprint for building a more sustainable future across the globe. There are totally 17 goals in SDGs, addressing issues in economic, environmental, and social areas. These goals are interrelated and expected to be fulfilled by 2030 (2015). PM2.5 emissions have an adverse effect on public health (Yin, Pizzol, and Xu 2017). At present, many countries attach great importance to inhalable total suspended particulates that cannot be blocked by the human upper respiratory tract, especially inhalable aerosol particles with sizes less than 2.5 μm. It is necessary to arouse the stakeholder
Conclusions
In this paper, an integrated analytical framework targeting sustainability analysis of supply chain management through particulate matter emissions prediction has been proposed. Specifically, we firstly collect and preprocess the particulate matter emission data and supply chain statistics in Beijing of China, then we conduct correlation analysis between air pollutant concentrations and supply chain statistics to uncover the intrinsic associations. Secondly, we combine the boosting ensemble meta-algorithm and neural network method for particulate matter concentration forecasting. The boosting algorithm is used to train a set of individual subnets to improve the efficiency of prediction, and afterwards, the output results of individual subnets are weighted for acquiring the final prediction results. Finally, we perform an analysis on sustainability between air pollutant emissions and supply chain statistics after obtaining the particulate matter prediction results.
This analytical framework is able to perform sustainability analysis with accuracy and provide reliable insights for the legislation and implementation of green supply chain management. The forecasting of particulate matter emissions conveys basis and justifications for policy-makers to enforce the environmental protection measures associated with air pollution and design efficient policies with the aim of improving the air pollutant management of the supply chain, which will further lead to an improved public health environment, especially in economically developed areas. Because this study mainly targets the prediction of PM2.5 and PM10 concentrations, it cannot comprehensively reveal the geographical distribution of air pollution. There are various factors that affect air quality, and a regression model can be established to predict PM2.5 and PM10 distributions more comprehensively and to determine the key influential factors, which would be important for formulating economic development strategies and implementing environmental protection mechanisms.