CN-122022510-A - Integrated management method and system for green low-carbon supply chain

CN122022510ACN 122022510 ACN122022510 ACN 122022510ACN-122022510-A

Abstract

The application discloses an integrated management method and system for a green low-carbon supply chain, wherein the method comprises the steps of collecting and integrating multi-source unstructured text data of the whole chain of the supply chain, carrying out green performance evaluation on suppliers in the supply chain based on the multi-source unstructured text data of a disadvantaged reason text, generating an evaluation report based on a green performance evaluation result, calculating full life cycle carbon footprint information of products or services based on the multi-source unstructured text data of the disadvantaged reason text, carrying out tracing and visual display, and establishing a multi-objective optimization model and outputting optimal decisions based on the evaluation report of the disadvantaged reason text and the full life cycle carbon footprint information of the disadvantaged reason text. By utilizing the scheme provided by the application, the objective and quantitative evaluation of the green performance of the supplier is realized, and the full life cycle carbon footprint of the visualized product is precisely traced. Meanwhile, the optimal decision is output by means of the multi-objective optimization model, and the power supply chain can be comprehensively assisted to realize green and intelligent upgrading.

Inventors

TIAN YE
LI YI
SHI QI
Xiang Sijing
SONG BO
Yang Dijiao
AN MING

Assignees

中国大唐集团绿色低碳发展有限公司

Dates

Publication Date: 20260512
Application Date: 20251208

Claims (10)

1. The integrated management method for the green low-carbon supply chain is characterized by comprising the following steps of: Collecting and integrating multi-source unstructured text data of a whole chain of a supply chain; performing green performance evaluation on suppliers in a supply chain based on the multi-source unstructured text data, and generating an evaluation report based on green performance evaluation results; Calculating full life cycle carbon footprint information of a product or service based on the multi-source unstructured text data, and performing traceability and visual display; And based on the evaluation report and the full life cycle carbon footprint information, establishing a multi-objective optimization model and outputting an optimal decision.
2. The green low-carbon supply chain integrated management method of claim 1, wherein the multi-source unstructured text data comprises internal unstructured text data and external unstructured text data; The internal unstructured text data comprises product data and raw material data uploaded by enterprises; the external unstructured text data includes ESG reports provided by suppliers, carbon accounting reports, enterprise environmental penalty news data obtained from internet public channels, industry green index data, government environmental protection promulgation data.
3. The green low-carbon supply chain integrated management method according to claim 1 or 2, wherein in the process of green performance evaluation of suppliers in a supply chain based on the multi-source unstructured text data, the following steps are performed: constructing a green performance evaluation large language model; Extracting key green performance indexes based on multi-source unstructured text data through a green performance evaluation large language model, and performing cross verification on the key green performance indexes and supplier filling data; and quantitatively scoring the suppliers based on the cross-validated supplier filling data.
4. The integrated management method of a green low-carbon supply chain according to claim 3, wherein in the process of extracting key green performance indicators based on multi-source unstructured text data through a green performance evaluation large language model and cross-verifying the key green performance indicators and supplier filling data, the following steps are performed: Preprocessing the multi-source unstructured text data, wherein the preprocessing comprises cleaning and normalization processing; Matching the preprocessed multi-source unstructured text data with a pre-trained green field word stock through a green performance evaluation large language model, extracting key green performance indexes of suppliers from the preprocessed multi-source unstructured text data, and associating corresponding time dimensions and calculation calibers; Classifying and mapping the extracted key green performance indexes according to preset categories to enable the key green performance indexes to correspond to index dimensions in the supplier filling data; and performing cross verification including data dimension alignment, numerical consistency verification and logic relevance verification on the supplier filling data through the classified mapped key green performance indexes.
5. The integrated management method of a green low-carbon supply chain according to claim 3, wherein in the process of quantitatively scoring the suppliers based on the cross-validated supplier filling data, the following steps are performed: constructing a hierarchical predefined green low-carbon evaluation index system comprising primary indexes and secondary indexes, and setting weights of the indexes of each level and industry standard pole values; Determining input data of each secondary index based on the cross-validated vendor fill data; based on the input data of each secondary index, obtaining a quantization score corresponding to each secondary index; based on the quantized scores and the corresponding weights of the secondary indexes, performing upward aggregation calculation to obtain corresponding primary index scores; and (5) obtaining the green performance score of the provider by carrying out weighted summation on all the primary indexes.
6. The integrated management method of a green low-carbon supply chain according to claim 5, wherein the secondary index includes a quantitative type secondary index and a qualitative type secondary index; in the process of obtaining the quantization score corresponding to the quantitative type secondary index, the following calculation formula is adopted: , wherein, Is the score of the jth secondary index under the ith primary index, Is the input data of the secondary index, Is an industry standard value; in the process of obtaining the quantized score corresponding to the secondary index of the qualitative type, the following steps are executed: Carrying out semantic matching on the input data of the secondary index and a preset evaluation standard to obtain an evaluation grade corresponding to the input data of the secondary index; and determining the quantized score corresponding to the secondary index in the score interval corresponding to the evaluation grade.
7. The green low-carbon supply chain integrated management method of claim 5, wherein the assessment report includes dominance data, disadvantaged data, and improvement policies; in the process of generating the dominance data, the following steps are performed: traversing all the primary index scores and the secondary index scores of the suppliers, and judging the index with the score larger than or equal to the corresponding preset high-score threshold value as the dominant index; retrieving the cross-validated supplier filling data corresponding to each advantage index and forming a corresponding advantage description text; In generating the disadvantaged data, the following steps are performed: Traversing all the primary index scores and the secondary index scores of the suppliers, and judging the indexes with the scores smaller than or equal to the corresponding preset high-score thresholds as inferior indexes; Calling the cross-validated provider report data corresponding to each disadvantage index, and forming a corresponding disadvantage reason text; In generating the improved policy, the following steps are performed: invoking a preset knowledge base, wherein the knowledge base comprises target rod cases with different industries and different index dimensions; Mapping the cross-validated supplier filling data corresponding to each disadvantaged index into a knowledge base, and searching and matching the same industry winning pole case; An improvement strategy is generated based on matching pole cases in the same industry.
8. The integrated management method of a green low-carbon supply chain according to claim 1, wherein in the process of establishing a multi-objective optimization model and outputting an optimal decision, the following steps are performed: screening suppliers meeting the preset green performance threshold condition based on the green performance evaluation result, thereby forming a qualified supplier set; constructing a multi-objective optimization model aiming at the qualified provider set, wherein the expression of the multi-objective optimization model is as follows: , wherein, For the total operating cost of the ith qualified provider, For the amount of product purchased from the ith qualified provider, For the unit product price of the ith qualified supplier, For a freight per unit distance per unit product in transportation mode k, For the distance of the ith qualified provider, For the total carbon emissions of the ith qualified supplier, Is the carbon emission amount of the unit product, Carbon emissions per unit distance per unit product in transportation mode k; Calculating the maximum total operation cost and the maximum total carbon emission corresponding to the qualified supplier set based on the multi-objective optimization model; And (3) based on the maximum total operation cost and the maximum total carbon emission, minimizing the comprehensive performance targets corresponding to the qualified suppliers, and obtaining the optimal decision of the qualified suppliers.
9. The integrated management method of a green low-carbon supply chain according to claim 8, wherein the expression of the comprehensive performance objective corresponding to each qualified provider is: , wherein, For the composite performance objective corresponding to the ith qualified provider, For a maximum total operating cost corresponding to a set of qualified suppliers, For a maximum total carbon emission corresponding to a qualified supplier set, For the total operating cost of the ith qualified provider, For the total carbon emissions of the ith qualified supplier, For cost savings/carbon emission reduction weight.
10. A green low-carbon supply chain integrated management system, characterized in that the green low-carbon supply chain integrated management is performed by using the green low-carbon supply chain integrated management method according to any one of claims 1 to 9, the system comprising: the data acquisition module is used for acquiring and integrating multi-source unstructured text data of a full chain of a supply chain; The green performance evaluation module is used for performing green performance evaluation on suppliers in a supply chain based on the multi-source unstructured text data and generating an evaluation report based on a green performance evaluation result; the carbon footprint information acquisition module is used for calculating the full life cycle carbon footprint information of the product or service based on the multi-source unstructured text data and carrying out traceability and visual display; And the optimal decision output module is used for establishing a multi-objective optimization model and outputting an optimal decision based on the evaluation report and the full life cycle carbon footprint information.

Description

Integrated management method and system for green low-carbon supply chain Technical Field The present application relates generally to the field of supply chain management. More particularly, the application relates to an integrated management method and system for a green low-carbon supply chain. Background In the global "two carbon" (carbon peak, carbon neutralization) strategic context, manufacturing supply chain management is facing a profound paradigm shift, namely a shift from local, static, environmental compliance measures to full life cycle, dynamic, precise management of carbon footprint. Currently, although advanced manufacturing industries such as automobiles and electronics have been tried on, in the actual landing process, the existing supply chain management technical scheme still has difficulty in meeting the transformation requirement. The prior art has the following significant drawbacks when dealing with full life cycle management under the "two carbon" goal: Firstly, the data circulation is limited, and the deep analysis and verification capability of multi-source data is lacking, wherein the prior art realizes the integration of partial data but is mainly limited to logistics inventory data in enterprises or directly related to the enterprises. The data standards of all links of the supply chain are different, so that a cross-organization information barrier is formed, and the carbon footprint is difficult to realize the through and the traceability of the whole chain. Especially for green low-carbon performance evaluation of suppliers, the prior art relies on static data self-reported by the suppliers, lacks the capability of utilizing external public information (such as policy supervision and industry database) to carry out automatic verification, deep analysis and intelligent evaluation, and is difficult to guarantee the authenticity and accuracy of the data. Secondly, the targets are unbalanced, and a dynamic collaborative optimization mechanism of economic and environmental targets is lacked, so that the prior art is still a collaborative tool with economic efficiency (such as response speed and performance rate) as a core. Environmental objectives are only one type of auxiliary monitoring indicator or soft constraint in the system, and lack a mechanism to incorporate carbon emissions as a core optimization objective into the system schedule. Under the scene of multi-objective conflict of low carbon, low cost and the like, the existing system lacks the capability of dynamic balance and collaborative optimization, and the balance of economic benefit and environmental benefit is difficult to realize in core links of resource scheduling, supplier selection and the like. Thirdly, the decision intelligentization degree is low, prospective low-carbon decision suggestions cannot be generated, traditional life cycle assessment and the existing optimization model are highly dependent on manually set rules and parameters, and massive and multi-source heterogeneous data are difficult to process. The existing system cannot perform self-adaptive learning based on macroscopic policy, industry dynamic and microscopic operation data, so that comprehensive improvement suggestions with environmental benefits and economic feasibility cannot be provided for managers. The decision process still relies heavily on manual experience, lacking accurate and prospective intelligent decision support. In view of the foregoing, it is desirable to provide a green low-carbon supply chain integrated management scheme to solve the above problems and dynamically balance the economic cost and environmental impact. Disclosure of Invention In order to solve at least one or more of the technical problems mentioned above, the present application proposes, in various aspects, a green low-carbon supply chain integrated management scheme. In a first aspect, the application provides a green low-carbon supply chain integrated management method, which comprises the steps of collecting and integrating multi-source unstructured text data of a whole supply chain, carrying out green performance evaluation on suppliers in the supply chain based on the multi-source unstructured text data, generating an evaluation report based on a green performance evaluation result, calculating whole life cycle carbon footprint information of a product or service based on the multi-source unstructured text data, carrying out traceability and visual display, and establishing a multi-objective optimization model and outputting an optimal decision based on the evaluation report and the whole life cycle carbon footprint information. In some embodiments, the multi-source unstructured text data includes internal unstructured text data including product data and raw material data uploaded by an enterprise and external unstructured text data including ESG reports provided by a provider, carbon accounting reports, and enterprise environmental penalty news d