Open AccessArticle

Resilience Evaluation of the Forest Products Platform Supply Chain Based on Artificial Intelligence and Extension Theory

Lin Lu

¹,

Ping Long

^1,* and

Xiaochun Luo

^1,2

School of Economics and Management, Guangxi Normal University, Guilin 541004, China

College of Economics and Management, Nanjing University of Aeronautics and Astronautics, Nanjing 210016, China

Author to whom correspondence should be addressed.

Forests 2024, 15(12), 2180; https://doi.org/10.3390/f15122180

Submission received: 1 November 2024 / Revised: 27 November 2024 / Accepted: 4 December 2024 / Published: 11 December 2024

(This article belongs to the Section Wood Science and Forest Products)

Download

Browse Figure

Versions Notes

Abstract

Forestry has a profound impact on environmental protection, economic development, and social welfare. With the improvement of global environmental protection awareness, the construction of platform supply chain of forest products has become the core driving force to promote sustainable development of forestry. Studying the resilience of supply chain of platform of forest products is of great importance to solve the contradiction between economic development and natural ecosystem protection. However, the existing resilience evaluation methods are not suitable for the dynamic and complex performance evaluation of the current forest products platform supply chain. Therefore, in order to make up for this shortcoming, this paper evaluates and analyzes the supply chain resilience of the forest products platform based on AI recommendation and extension theory. Firstly, this paper combined the characteristics of forest products and used AI recommendation technology to build a forest products platform supply chain resilience performance evaluation index system. Secondly, the AHP method was used to calculate the index weight, and the resilience evaluation model of the platform supply chain of forest products was constructed. Finally, in order to ensure the authenticity and credibility of the evaluation results, three practical cases were analyzed to illustrate the resilience level of the platform supply chain of forest products, and the effectiveness of the application of AI recommendation and extension theory in the resilience performance evaluation of forest products platform supply chains was verified. The scientific value of this paper is that it provides a new idea and a new method for the resilience performance evaluation of the forest products platform supply chain and makes theoretical and practical contributions to the fruitful application of AI recommendation in the supply chain field. In addition, this study also provides a new practical guideline for protecting the natural environment and realizing the sustainable development of forestry.

Keywords:

forestry sustainable development; forest products platform supply chain; resilience evaluation; artificial intelligence; extension theory

1. Introduction

At present, we are in the background of economic globalization, diversified market demand [1], technological progress, and abrupt climate change [2,3]. With the deepening of economic globalization, the production, processing, and distribution of forest products involve many regions and countries, and the forest products supply chain network becomes more complex [4], which leads to high vulnerability of the forest products supply chain. Secondly, due to the diversification of consumer demand for forest products, there is an urgent need to enhance the flexibility of the forest products supply chain in order to respond quickly to market changes. Then, technological progress provides new opportunities for the operation and development of the supply chain, and the application of technology can help optimize the management of the forest products supply chain and improve operational efficiency. Finally, climate change has led to frequent extreme weather [5] and increased uncertainty in natural ecosystems [6], which directly threatens the output, transportation, and sale of forest products and increases the operational instability of supply chains. Therefore, modern technology is needed to build more resilient supply chains for forest products [7]. Supply chain resilience refers to the ability of the supply chain to make plans quickly and maintain normal operations by solving problems in the face of uncertain risks [8,9]. The essence of supply chain resilience is an adaptive cycle process [10]. Through systematic reorganization, development, and adaptation, the supply chain can maintain normal operation in a disturbed environment. Supply chain resilience involves different stages of the supply chain from risk occurrence to recovery: preparation, response, recovery, and growth stages [11,12]. The content of supply chain resilience is different in different stages. In the preparation stage, a supply chain with resilience can accurately perceive risks before they occur and enhance the robustness of the chain. In the response phase, a resilient supply chain can respond in time and ensure the continuity of supply chain operations before the crisis expands. In the recovery and growth phase, a resilient supply chain can accumulate and absorb experience after disruption and transform it into risk perception and response capacity through learning and innovation [13]. Forest is one of the most important ecosystems on earth, and its protection and sustainable management are crucial to maintaining biodiversity, water cycle, and soil protection [14]. Studying the resilience of the forest products supply chain can help improve the supply chain’s ability to cope with emergencies and ensure supply chain security and timely supply of products. It is also helpful for enterprises to achieve benign development of economic and environmental benefits [15]. In conclusion, the study of the resilience of the forest products supply chain is not only vital for the healthy development of forestry, but also has great significance for ensuring economic security and promoting the sustainable development of forestry. Only with resilience can the supply chain of forest products form a healthy ecology of sustainable development. By improving the resilience of the forest products supply chain, we can build a more stable supply chain system and lay a solid foundation for meeting future challenges.

At the same time, the development trend of the supply chain platform leads the optimization and innovation of the supply chain model and promotes the transformation of the traditional solidified forest product supply chain into a dynamic model. The platform improves the information transparency of the forest product supply chain, and the platform integrates the data of all links of the supply chain, which greatly improves the cooperation efficiency of the supply chain. Effectively reduce the waste of products and promote the sustainable development of the supply chain.

In addition, the application of artificial intelligence in the supply chain has increasingly become a popular trend [16], and artificial intelligence can optimize supply chain management and improve supply chain resilience. Unfortunately, research and understanding about the resilience of forestry economy are not enough, and the literature on the identification of resilience indicators and the measurement of the resilience level is relatively limited. Therefore, this paper identifies the following three research objectives:

(1): Combined with the existing research results, the AI recommendation system was used to determine the relevant indicators of the supply chain resilience assessment of the forestproducts platform;
(2): The weight of each index was determined by analytic hierarchy Process (AHP), and the toughness evaluation model of the supply chain of forest products platform was constructed by extension theory;
(3): Evaluated the supply chain resilience level of different types of forest products platforms and verified the effectiveness of the application of artificial intelligence and extension theory in the supply chain resilience evaluation.

This study has three major contributions to the research field of forestry economic development. First, we discussed an under-researched issue overlooked in previous studies: How to assess the resilience level of the forest products platform supply chain. Considering the important position of forestry economy in the world economy and the complexity of the forest products platform supply chain, we combined artificial intelligence and extension evaluation methods to put forward a new method for evaluating the supply chain resilience of the forest products platform. Secondly, combined with the characteristics of forest products, the paper created an index system from five dimensions: production resilience, product supply resilience, economic resilience, logistics resilience, and risk management and resilience. The new evaluation index added ensures the comprehensiveness of the evaluation system. Finally, an example was used to evaluate the resilience level of the different types of forest products platform supply chains, and the validity of the model was verified. This provides valuable insights and practical guidance for improving the resilience of the forestry economy, protecting natural ecosystems, and applying artificial intelligence to sustainable supply chain development.

2. Literature Review

2.1. Forest Products Platform Supply Chain

Forest products mainly include forest products, forest by-products, forest agricultural products, etc. With the increase of forest products deep processing and extension products, it has become the main way of forest economic development. Existing scholars’ research on the forest products supply chain focuses on three aspects: supply chain model, resource integration, and relationship coordination [17]. Foreign experts and scholars believe that the supply chain can be divided into four basic models, namely, dumbbell type, T-type, symmetrical type, and comprehensive type [18], and explain the essence of supply chain management [19]. From the perspective of forestry resource integration, foreign scholars believe that the establishment of the forest product supply chain is to improve the overall efficiency of the supply chain, so it is necessary to supervise each link. From the perspective of supply chain relationship coordination, Mighell first defined the concept of vertical supply chain coordination [20]. Some scholars also believe that focusing on coordinating the benefit distribution relationship between supply chain producers and processors can enable enterprises to better improve the efficiency of the forest products supply chain [21].

The forest products platform supply chain mainly consists of raw material suppliers, platform (warehousing and logistics), and consumers, as shown in Figure 1. The supply chain of forest products platform is divided into upstream, midstream, and downstream. The upstream of the supply chain is located at the starting point of the supply chain of the forest products platform, providing products for the entire supply chain, while the middle reaches of the supply chain are controlled by the platform and provide intermediate services for the entire supply chain, while the downstream of the supply chain of the forest products platform directly contacts consumers [22].

2.2. Supply Chain Resilience Performance Evaluation Indicators and Evaluation Methods

From the perspective of resilience evaluation indicators, production resilience [23], purchasing resilience, economic resilience [24], and inventory resilience are mentioned most in current studies [25]. Meanwhile, flexibility, redundancy, collaboration, and agility are also the focus of researchers. As for the screening and judgment of indicators, some scholars adopted the DEMATEL method to identify and analyze the causality of various indicators in the supply chain and help determine which indicators have the greatest impact on performance, thus providing a scientific basis for the selection of indicators [26]. Another study proposes a group consensus mathematical model for the selection of supply chain indicators, which adopts the Delphi method to select performance indicators and uses the feedback mechanism based on the group analytic hierarchy process. The effective use of this model promotes the sustainable growth and development of the supply chain [27]. It is not difficult to see from the literature analysis that the current supply chain performance evaluation index selection methods are mature and have been applied to real life, but they are complicated and cost a lot of time. In addition, the current supply chain resilience evaluation index system lacks comprehensiveness and fails to cover the whole link of supply chain operation.

From the perspective of toughness evaluation methods, the most commonly used method is to measure and evaluate the toughness level of the supply chain by the entropy value method [28,29]. In addition, some scholars believe that resilience is a multi-dimensional ability and that multiple indicators should be considered in resilience assessment. Therefore, they developed an agent-based resilience analysis framework to judge the resilience of a system after damage and evaluate the resilience of the system [30]. Some scholars designed a restorable model using Bayesian networks, adjusted the parameters of the learning network model by collecting data samples, analyzed the relationship between various indicators and system toughness, and identified the key points for toughness improvement [31]. Martin and Gardiner [32] proposed a sensitivity algorithm, which constructed a regional economic resilience index by selecting one or two representative core variables to measure the level of economic resilience. It is not difficult to see that the existing toughness evaluation methods are abundant. However, the more significant problem of the above evaluation methods is the lack of systematic and objective defects, which may affect the accuracy of the evaluation results.

Through a review of these studies, it can be seen that the existing supply chain performance evaluation methods are abundant. In addition to common analysis methods, existing studies have innovatively evaluated supply chain performance through modeling. However, the problems of the above performance evaluation methods are the lack of dynamism and prominent subjectivity. The feedback mechanism of the existing methods for the performance evaluation results is not perfect, and they cannot timely reflect the dynamic changes of the enterprise.

2.3. Research on the Application of the Artificial Intelligence Recommendation and Extension Theory in Supply Chain Resilience

With the increasingly prominent role of the artificial intelligence recommendation in supply chain management, more and more enterprises take artificial intelligence as a way to improve the sustainable development ability and competitiveness of the supply chain [33]. The application of AI recommendations to supply chain resilience is an emerging and rapidly growing field, with the core being the use of algorithms and predictive models to enhance the responsiveness of supply chains to market volatility and uncertainty factors. In fact, some scholars have discussed it in combination with the actual situation. Some scholars have pointed out that artificial intelligence (AI) is a mechanism to improve supply chain resilience by developing business continuity capabilities [34]. Other scholars have examined the application of AI in enterprises, and studies have shown that AI improves the resilience of supply chains by improving supply chain visibility, risk resilience, procurement, and distribution capabilities. Specifically, AI can predict demand fluctuations by analyzing historical data, market trends, and external factors (such as economic indicators, seasonal changes, etc.) to optimize supply chain inventory management and production planning [35]. Artificial intelligence can improve the transparency of the supply chain, track the entire process of products from production to delivery, help consumers understand the source and quality of products, and help enterprises deal with information overload [36] and improve decision-making efficiency [37]. Artificial intelligence provides innovative methods for the development of the supply chain [38]. Some studies have pointed out that artificial intelligence has been applied to the production process of forest products [39], and artificial intelligence can identify and evaluate potential risks in the supply chain [40], including natural disasters and supplier stability, and help enterprises make emergency plans.

In view of the application of the extension theory in supply chain toughness, there are few existing studies, but a few scholars have applied the matter-element extension model to risk assessment and effectively prevented accidents. The effectiveness of practical application provides strong evidence for the reliability of the model [41]. Combined with practical cases, it is not difficult to see that the application of the extension theory in risk prevention and control can help managers assess the comprehensive impact of various risk factors and determine the key areas of risk management, which is conducive to the improvement of enterprise cost management and operational efficiency. The application of the extension theory to performance evaluation can bring some new perspectives and improvements to the traditional performance evaluation methods. Extension theory can better deal with the uncertainty and fuzziness in performance evaluation, which makes the evaluation results closer to the actual situation. Secondly, extension allows the evaluation criteria to be dynamically adjusted during the evaluation process, which makes the performance evaluation system more flexible [42]. Finally, the extension method combines quantitative and qualitative indicators to realize multi-dimensional and multi-level comprehensive evaluation and improves the comprehensiveness and accuracy of performance evaluation. Compared with traditional evaluation methods, although the extension theory has the above advantages, it still has some defects. First of all, it is difficult to construct an extension comprehensive evaluation model with relatively high complexity, which may be difficult for non-professionals in practical applications. Secondly, in the process of scoring indicators, it is often necessary to rely on the experience and judgment of experts, which may cause the evaluation result to be affected by personal factors. Finally, for the large-scale performance evaluation system, the extension method may lead to a large amount of calculations and an increase in time cost.

In summary, many cases have shown that the application of the artificial intelligence recommendation and the extension theory in supply chain resilience is feasible and effective. Therefore, the application of the artificial intelligence recommendation and the extension theory in the evaluation of supply chain flexibility performance is conducive to the sustainable development of the supply chain.

3. AI Recommended the Design of Resilience Evaluation Index System of Forest Products Platform Supply Chain

At present, there are abundant research literatures in the field of supply chain resilience evaluation index system construction in the theoretical circle, but most of them stay in broad perspectives such as organizational structure and system construction of performance management, and they cannot study the resilience evaluation index system of the forest products platform supply chain from a practical perspective, which makes it difficult for the industry to find a relevant reference basis in practice. Therefore, according to the design requirements of the resilience evaluation index of the supply chain of the forest products platform, this paper adopts the method of classification and subitem to construct the evaluation index. The main part of the forest products platform supply chain resilience evaluation index system constructed in this paper is roughly composed of five levels, namely, production resilience, product supply resilience, economic resilience, logistics resilience, and risk management and recovery ability. Among them, the above five levels are the criterion level, and the index level is also subdivided under the criterion level. They constitute the resilience performance evaluation index system of the forest products platform supply chain. In addition, we have also explained the data sources referenced by the evaluation of each indicator, as shown in Table 1.

(1): Production toughness index design

The production resilience index can be evaluated from the perspectives of resource acquisition, production technology, and value creation. Since the supply chain of the forest products platform has multiple characteristics, which have both commercial and environmental protection attributes, the production resilience of the supply chain of the forest products platform needs to be evaluated from the dual perspectives of self-benefit and social benefit. In addition, technology, as a key factor affecting productivity, should also be included in the assessment of production toughness, which makes the evaluation indicators of production toughness more diversified.

(2): Product supply toughness index design

The product supply resilience dimension focuses on product supply and inventory, and selects four indicators: “supplier diversification”, “product quantity resilience”, “product supply efficiency”, and “inventory management”. “Supplier diversification” examines the balance of the number of suppliers in the supply chain, preventing the supply chain from relying on a single supply chain to provide products. “Product quantity resilience” assesses the ability of the forest products platform supply chain to adapt to changes in market demand. “product supply efficiency” looks not only at the speed at which products are supplied in the day-to-day operation of the supply chain, but also at the speed at which the supply chain can meet market demand in unexpected situations. “Inventory management” affects the resilience level of the supply chain to a certain extent. Too much inventory leads to the waste of resources and the increase of operating costs, while too little inventory is not conducive to the enterprise’s response to emergencies.

(3): Economic resilience index design

Economic resilience is mainly determined by four interacting subsystems of economic structure, labor, finance, and innovation management. Economic resilience can be analyzed from four aspects: capital, market, price, and output. Among them, economic resilience contains six evaluation indicators, namely, “capital accumulation rate”, “profit growth rate”, “market potential”, “foreign trade dependence”, “price resilience”, and “innovation output”.

(4): Logistics toughness index design

The level of logistics resilience mainly focuses on two aspects: transportation mode and information transmission, including eight specific indicators. The transportation aspect involves “diversity of transportation modes” and “transportation reliability”, which measures the capacity and efficiency of the supply chain to use transportation means. In terms of information communication, the two indicators of “information timeliness rate” and “delivery timeliness” are examined to measure the speed of information transmission and the accuracy of product delivery in the supply chain. Finally, it also includes four indicators: “storage resilience”, “distribution resilience”, “logistics network”, and “requirements response speed”.

(5): Risk management and resilience index design

This level covers three indicators, namely, “risk identification and assessment”, “risk management strategy”, and “recovery speed and effectiveness”, which assess the risk management and recovery ability of the supply chain from different aspects to ensure the integrity of the resilience evaluation system.

In short, the index system can reflect the overall performance evaluation objectives, and at the same time, it can take into account different aspects of the supply chain of the forest products platform.

4. Methods

Extension is an original cross-sectional discipline proposed by Chinese scholar Cai Wen in 1983. It uses formal models to study the possibility of expansion of things and the rules and methods of innovation and is used to solve contradictory problems [43]. Primitive theory, extension set theory, and extension logic are the three foundations of extension theory [44]. The extension theory and extension innovation methods are combined with specific fields to form a number of operable methods, and their application in specific fields is called extension engineering. Extension theory, extension innovation method, and extension engineering constitute extension theory [45]. Its research goal is to formalize contradiction problems so that computers can understand them, that is, to intellectualize contradiction problems, and to study and solve contradiction problems by combining qualitative analysis and quantitative calculation [46].

The concept of matter elements scientifically reflects the relationship between quality and quantity, and it is more scientific to describe the process of changes in objective things [47]. Matter-element theory enables us to describe various possibilities of transaction changes through mathematical language, especially the causal relationship between behavior chains and things [48], so that we can not only use the causal relationship between things to formulate solutions to problems, but can also use the transformation of material elements to study the changes of things. The concrete steps of building the matter-element extension model are as follows:

Step 1: Determine the matter-element matrix.

Matter-element theory is used to describe the three elements of a thing: the ordered triplet composed of the thing M, the feature C, and the magnitude V.

R = (M, C, V)

is called the basic element of a thing, that is, the matter element. According to the definition and characteristics of matter elements, the matter-element matrix of a thing with multiple characteristics is denoted as:

R = [M, C, V] = [\begin{matrix} M & C_{1} & v_{1} \\ C_{2} & v_{2} \\ ⋮ & ⋮ \\ C_{n} & v_{n} \end{matrix}]

(1)

note: n is the number of features of things, and the quantity value corresponding to each feature is v₁, v₂,…, v_n.

Step 2: The matter-element matrix of classical domain and section domain.

Classical domain matter-element matrix: The classical domain is a classification of selected evaluation objects, indicators, and standard value intervals of indicators, that is, the range of values of indicators corresponding to the level of evaluation. The form is as follows:

\begin{array}{l} R_{i} = [N_{i}, C_{n}, V_{in}] = [\begin{matrix} N i & C_{1} & v_{i 1} \\ C_{2} & v_{i 2} \\ ⋮ & ⋮ \\ C_{n} & v_{i n} \end{matrix}] \\ = [\begin{matrix} N_{i} & C_{1} & (a_{i 1}, b_{i 1}) \\ C_{2} & (a_{i 2}, b_{i 2}) \\ ⋮ & ⋮ \\ C n & (a i n, b i n) \end{matrix}] \end{array}

(2)

note: R_i is the classical domain object element, N_i represents i performance evaluation levels, C₁, C₂,…, C_n represent evaluation indicators, and v_i1, v_i2,…, v_in is the classical domain of matter-element matrix, that is, the value interval of the evaluation level i of things to be evaluated.

Section domain matter-element matrix: Section domain refers to all value ranges of selected evaluation objects, indicators, and standard values of indicators, and is the union of all value ranges in the classical domain. Section domain matter-element matrix is a matter-element matrix composed of the things to be evaluated, the characteristics of the things, and the corresponding expanded value range of the characteristics, denoted as Rp, in the following form:

\begin{array}{l} R_{p} = [P, C_{n}, V_{P}] = [\begin{matrix} P & C_{1} & v_{P 1} \\ C_{2} & v_{P 2} \\ ⋮ & ⋮ \\ C_{n} & v_{P n} \end{matrix}] \\ = [\begin{matrix} P & C_{1} & (a_{P 1}, b_{P 1}) \\ C_{2} & (a_{P 2}, b_{P 2}) \\ ⋮ & ⋮ \\ C_{n} & (a_{P n}, b_{P n}) \end{matrix}] \end{array}

(3)

note: R_p is the section domain object element, P represents the whole composed of evaluation levels, C_n represents each evaluation index, and v_p1, v_p2, …, v_pn is the section field of the matter-element matrix, representing all the values of all evaluation levels.

Step 3: Determination of the weight coefficient.

The determination of the weight of each index has a crucial impact on the subsequent calculation. The final result will be different depending on how the weight is measured. In existing studies, the analytic hierarchy process is the most commonly used method [49,50], using the advantages of this method in determining weights, the weights of evaluation indicators can be better determined, and finally, the importance of each indicator can be fully reflected.

First of all, under the guidance of experts, the Delphi method is used to determine the relative degree, construct the judgment matrix, and make a pairwise comparison of the elements at the same level. The value of the judgment matrix reflects the importance of each element, usually using the Santy 1–9 scale method to build the matrix. Then, the expert scoring method is used to evaluate the relative importance of each indicator, summarize the expert evaluation, and finally calculate the weight value of each level of indicators and the maximum feature root of the judgment matrix, and carry out the consistency test. The formula for calculating the consistency index

C I

is as follows:

C I = \frac{λ \max - n}{n - 1}

(4)

where

λ \max

is the maximum eigenvalue of the judgment matrix. To calculate the consistency rate

C R

, the formula is as follows:

C R = C I / R I

(5)

Step 4: Correlation degree function and evaluation level.

Correlation degree is a quantitative tool that can describe the membership relationship among things. The correlation degree is used to express the degree of correlation between the object element to be evaluated and each evaluation level in a numerical way. The value range is in the whole real number interval, and its function is equivalent to the membership degree in the fuzzy mathematical model. Therefore, the correlation function is an evaluation tool for a certain indicator in the performance evaluation model, and its essence is to reflect the degree of fit between the two. The greater the correlation value of a certain level, the higher the membership degree of the object to be evaluated to that level. The form is as follows:

\begin{matrix} k_{e} (v_{i}) = \{\begin{matrix} \frac{ρ [v_{i} (e), V_{e j}]}{ρ [v_{i} (e), V_{p i}] - ρ [v_{i} (e), V_{e j}]} \\ (ρ [v_{i} (e), V_{p i}] - ρ [v_{i} (e), V_{e j}] \neq 0) \\ - ρ [v_{i} (e), V_{e j}] - 1 \\ (ρ [v_{i} (e), V_{p i}] - ρ [v_{i} (e), V_{e j}] = 0) \end{matrix} \\ ρ [v_{i} (e), V_{e j}] = |v_{i} - \frac{1}{2} (a_{e j} + b_{e j})| - \frac{1}{2} (b_{e j} - a_{e j}), \\ ρ [v_{i} (e), V_{p i}] = |v_{i} - \frac{1}{2} (a_{p i} + b_{p i})| - \frac{1}{2} (b_{p i} - a_{p i}) \end{matrix}

(6)

note:

k_{e} (v_{i})

denotes the value of the correlation function between the i-th indicator and the j-th level in the supply chain performance evaluation (e = 1, 2, …, s),

ρ (v_{i}, V_{e j})

denotes the distance between the value of the i-th indicator of the object element to be evaluated and the classical domain,

ρ (v_{i}, V_{p i})

denotes the distance between the value of the i-th indicator of the object element to be evaluated and the node domain, and

V_{pi}

denotes the range of values of the node domain of the i-th indicator (i = 1, 2, …, m).

According to the weight of the indicators

w_{i}

, the comprehensive correlation degree of each indicator can be calculated, that is, the degree of belonging of the indicators to be evaluated to each evaluation level, which can be expressed as:

k_{e} (p_{o}) = \sum_{i = 1}^{m} ω_{i} k_{e} (v_{i})

(7)

If the evaluation criteria are formulated for the decision of a single goal, it is mainly based on the maximum fuzzy membership degree. According to this standard, through the scientific evaluation of supply chain performance, the initial judgment of performance level can be obtained, namely:

k_{e} (p_{o}) = \max \{k_{e} (p_{o})\} (e = 1, 2, \dots, s)

(8)

5. Results

5.1. Identify the Classical Domain and the Section Domain

Combined with the types of forestry supply chain and platform supply chain models discussed in the existing research, this paper selects three different types of forest product platform supply chains with the help of the AI recommendation. A (Integrated Forest Products Platform Supply Chain): The supply chain takes inventory management and process optimization as the breakthrough, and relies on a professional global service network to provide raw material procurement, distribution, distribution, finance, and other services. It covers the whole process of the supply chain in procurement, production, sales, and service.

B (Supplier-led forest products platform supply chain): Suppliers use strong resource integration capabilities to build a supply chain, organize the procurement, production, and delivery of forestry products on a global scale, and provide services from product development, raw material procurement, production, assistance outsourcing, commissioned processing, process control, inventory management, quality testing, import and export agency, marketing, and other services, covering the entire supply chain, especially multiple value points in the production process. Promote the transformation and upgrading of traditional forestry enterprises.

C (E-commerce-led forest products platform supply chain): In line with the development trend of e-commerce, the supply chain builds a global transportation, warehousing, and distribution network based on the Internet and technology platform, provides customers with online and offline supply chain services for B2B and B2C, and realizes one-stop services from order management, source development, agent procurement, inventory management, international freight, logistics, and distribution. Enhance the ability of forestry enterprises to respond to customer needs and promote the flattening of trade channels.

We analyzed the toughness level of the above three kinds of forest product platform supply chains, so as to verify the operability of the supply chain toughness evaluation index system and extension goodness performance evaluation method constructed in this paper. In order to better reflect the resilience level of the supply chain, it is necessary to classify the resilience level of the supply chain of the forest products platform. In this paper, the resilience level is divided into five levels, which are excellent, good, average, poor, and very poor. The corresponding classic domains are [4, 5], [3, 4], [2, 3], [1, 2], and [0, 1]. The section domain is [0, 5].

According to the establishment of evaluation levels, the classic domain and section domain of the resilience performance evaluation of the forest products platform supply chain can be obtained.

Classic Domain:

Section Domain:

R_{j} = [\begin{matrix} N_{0} & N_{1} & N_{2} & N_{3} & N_{4} & N_{5} \\ C_{1} & 〈0, 1〉 & 〈1, 2〉 & 〈2, 3〉 & 〈3, 4〉 & 〈4, 5〉 \\ C 2 & 〈0, 1〉 & 〈1, 2〉 & 〈2, 3〉 & 〈3, 4〉 & 〈4, 5〉 \\ C_{3} & 〈0, 1〉 & 〈1, 2〉 & 〈2, 3〉 & 〈3, 4〉 & 〈4, 5〉 \\ C_{4} & 〈0, 1〉 & 〈1, 2〉 & 〈2, 3〉 & 〈3, 4〉 & 〈4, 5〉 \\ C_{5} & 〈0, 1〉 & 〈1, 2〉 & 〈2, 3〉 & 〈3, 4〉 & 〈4, 5〉 \\ C_{6} & 〈0, 1〉 & 〈1, 2〉 & 〈2, 3〉 & 〈3, 4〉 & 〈4, 5〉 \\ C_{7} & 〈0, 1〉 & 〈1, 2〉 & 〈2, 3〉 & 〈3, 4〉 & 〈4, 5〉 \\ C_{8} & 〈0, 1〉 & 〈1, 2〉 & 〈2, 3〉 & 〈3, 4〉 & 〈4, 5〉 \\ C_{9} & 〈0, 1〉 & 〈1, 2〉 & 〈2, 3〉 & 〈3, 4〉 & 〈4, 5〉 \\ C_{10} & 〈0, 1〉 & 〈1, 2〉 & 〈2, 3〉 & 〈3, 4〉 & 〈4, 5〉 \\ C_{11} & 〈0, 1〉 & 〈1, 2〉 & 〈2, 3〉 & 〈3, 4〉 & 〈4, 5〉 \\ C_{12} & 〈0, 1〉 & 〈1, 2〉 & 〈2, 3〉 & 〈3, 4〉 & 〈4, 5〉 \\ C_{13} & 〈0, 1〉 & 〈1, 2〉 & 〈2, 3〉 & 〈3, 4〉 & 〈4, 5〉 \\ C_{14} & 〈0, 1〉 & 〈1, 2〉 & 〈2, 3〉 & 〈3, 4〉 & 〈4, 5〉 \\ C_{15} & 〈0, 1〉 & 〈1, 2〉 & 〈2, 3〉 & 〈3, 4〉 & 〈4, 5〉 \\ C_{16} & 〈0, 1〉 & 〈1, 2〉 & 〈2, 3〉 & 〈3, 4〉 & 〈4, 5〉 \\ C_{17} & 〈0, 1〉 & 〈1, 2〉 & 〈2, 3〉 & 〈3, 4〉 & 〈4, 5〉 \\ C_{18} & 〈0, 1〉 & 〈1, 2〉 & 〈2, 3〉 & 〈3, 4〉 & 〈4, 5〉 \\ C_{19} & 〈0, 1〉 & 〈1, 2〉 & 〈2, 3〉 & 〈3, 4〉 & 〈4, 5〉 \\ C_{20} & 〈0, 1〉 & 〈1, 2〉 & 〈2, 3〉 & 〈3, 4〉 & 〈4, 5〉 \\ C_{21} & 〈0, 1〉 & 〈1, 2〉 & 〈2, 3〉 & 〈3, 4〉 & 〈4, 5〉 \\ C_{22} & 〈0, 1〉 & 〈1, 2〉 & 〈2, 3〉 & 〈3, 4〉 & 〈4, 5〉 \\ C_{23} & 〈0, 1〉 & 〈1, 2〉 & 〈2, 3〉 & 〈3, 4〉 & 〈4, 5〉 \\ C_{24} & 〈0, 1〉 & 〈1, 2〉 & 〈2, 3〉 & 〈3, 4〉 & 〈4, 5〉 \\ C_{25} & 〈0, 1〉 & 〈1, 2〉 & 〈2, 3〉 & 〈3, 4〉 & 〈4, 5〉 \\ C_{26} & 〈0, 1〉 & 〈1, 2〉 & 〈2, 3〉 & 〈3, 4〉 & 〈4, 5〉 \\ C_{27} & 〈0, 1〉 & 〈1, 2〉 & 〈2, 3〉 & 〈3, 4〉 & 〈4, 5〉 \\ C_{28} & 〈0, 1〉 & 〈1, 2〉 & 〈2, 3〉 & 〈3, 4〉 & 〈4, 5〉 \\ C_{29} & 〈0, 1〉 & 〈1, 2〉 & 〈2, 3〉 & 〈3, 4〉 & 〈4, 5〉 \end{matrix}]

R_{p} = [\begin{matrix} N_{p} & C_{1} & 〈0, 5〉 \\ C_{2} & 〈0, 5〉 \\ C_{3} & 〈0, 5〉 \\ C_{4} & 〈0, 5〉 \\ C_{5} & 〈0, 5〉 \\ C_{6} & 〈0, 5〉 \\ C_{7} & 〈0, 5〉 \\ C_{8} & 〈0, 5〉 \\ C_{9} & 〈0, 5〉 \\ C_{10} & 〈0, 5〉 \\ C_{11} & 〈0, 5〉 \\ C_{12} & 〈0, 5〉 \\ C_{13} & 〈0, 5〉 \\ C_{14} & 〈0, 5〉 \\ C_{15} & 〈0, 5〉 \\ C_{16} & 〈0, 5〉 \\ C_{17} & 〈0, 5〉 \\ C_{18} & 〈0, 5〉 \\ C_{19} & 〈0, 5〉 \\ C_{20} & 〈0, 5〉 \\ C_{21} & 〈0, 5〉 \\ C_{22} & 〈0, 5〉 \\ C_{23} & 〈0, 5〉 \\ C_{24} & 〈0, 5〉 \\ C_{25} & 〈0, 5〉 \\ C_{26} & 〈0, 5〉 \\ C_{27} & 〈0, 5〉 \\ C_{28} & 〈0, 5〉 \\ C_{29} & 〈0, 5〉 \end{matrix}]

5.2. Calculate Index Weight

The study adopted an analytic hierarchy process to calculate the weights of each evaluation index and invited 20 experts and scholars who have in-depth research on forestry and supply chain and related professional enterprise managers to form a research team. They are quite familiar with the supply chain system and forestry and have a deep understanding of the current industry status, and gave objective answers. In the form of a questionnaire, we invited them to score the evaluation indicators according to relevant data and score each resilience evaluation indicator according to its importance. The higher the score, the higher the importance, so as to determine the relative degree. The scoring results are shown in Table 2.

As can be seen from Table 2, A (Integrated forest products platform supply chain) has high scores for all the resilience evaluation indicators, especially in the three aspects of product supply resilience, logistics resilience, and risk management and resilience. According to the score of B (Supplier-led forest products platform supply chain), we can see that B has a good performance in logistics resilience and risk management and resilience, but a low score in production resilience, product supply resilience, and economic resilience, indicating that B has a low level of resilience in these three aspects. Finally, C (E-commerce-led forest products platform supply chain) has a high score in product supply resilience, economic resilience, logistics resilience, risk management and resilience, and a slightly low score in production resilience, indicating that there is still room for improvement in production resilience.

The analytic hierarchy process was used to calculate the weight of each evaluation index. According to the analytic hierarchy process, the judgment matrix of each level of indicators is constructed, the weight value of each level of indicators and the maximum feature root of the judgment matrix are calculated, and the consistency test is carried out. And according to the expert scoring table, the primary indicators and secondary indicators weights can be calculated, as shown in Table 3.

In Table 3, we can clearly see the importance and weight value of each resilience index. Among the five aspects of the resilience evaluation of the supply chain of the forest products platform, production resilience and logistics resilience occupy the largest proportion, indicating that these two aspects have a greater impact on the resilience level of the supply chain, followed by product supply resilience, risk management and resilience, and economic resilience. Specifically, in terms of the production resilience evaluation indicators, the four indicators of raw material availability, process resilience, quality control system, and technological innovation ability account for the highest proportion, indicating that these four indicators largely affect the production resilience level of the supply chain. Secondly, in terms of product supply resilience, supplier diversification accounts for the highest proportion, accounting for 46.58%, followed by inventory management, product supply efficiency, and the smallest proportion, which is product quantity resilience. Third, at the level of economic resilience, the four indicators of profit growth rate, market potential, price resilience, and innovation output account for a relatively large proportion, of which innovation output accounts for the largest proportion, 33.2%, indicating that innovation output has an important impact on economic resilience. Fourth, in terms of logistics resilience, information timeliness and delivery timeliness occupy an important position, the importance of the two is self-evident, and the two indicators involving transportation account for a relatively small proportion. Finally, in terms of risk management and resilience, we can see that recovery speed and effectiveness account for 53.9%, risk management strategies account for 29.72%, and risk identification and assessment account for 16.38%. Recovery speed and effectiveness have a significant impact on supply chain risk management and resilience.

5.3. Calculated Correlation Degree

According to Formula (6) and Table 2, the correlation function value of the secondary index can be calculated. Take C₁ of A as an example:

ρ (ν_{1}, ν_{11}) = |4.02 - \frac{1}{2} (0 + 1)| - \frac{1}{2} (1 - 0) = 3.02

ρ (ν_{1}, ν_{p 1}) = |4.02 - \frac{1}{2} (0 + 5)| - \frac{1}{2} (5 - 0) = - 0.98

\begin{array}{l} k_{1} (ν_{1}) = \frac{ρ (ν_{1}, ν_{11})}{ρ (ν_{1}, ν_{p 1}) - ρ (ν_{1}, ν_{11})} \\ = \frac{3.02}{- 0.98 - 3.02} = - 0.7550 \end{array}

Using the same method, the correlation degree of this indicator for other levels can also be calculated. In addition, the correlation degree of other secondary indicators relative to the five levels can also be calculated, and the specific results are shown in Table 4 and Table 5.

We can clearly see the level of each resilience index of the three supply chains and can easily make a horizontal comparison. The specific analysis is as follows:

A (Integrated forest products platform supply chain): In terms of production resilience, its raw materials availability, process resilience, and partner resilience are at an excellent level, and other indicators are at a good level. Secondly, in terms of product supply resilience, the supply chain’s supplier diversification, product quantity resilience, and product supply efficiency are excellent, and inventory management is at a good level. Third, in terms of economic resilience, its capital accumulation rate and profit growth rate are excellent, and other indicators are at a good level. Fourth, in terms of logistics resilience, all indicators are at an excellent level, which directly demonstrates the advantages of the supply chain in terms of logistics toughness. Finally, in terms of risk management and resilience, risk identification and assessment, recovery speed and effectiveness are at an excellent level, while risk management strategy is at a good level.

B (Supplier-led forest products platform supply chain): In terms of production resilience, its technological innovation ability, enterprise resilience culture, and partner resilience are at an average level, and other indicators are at a good level. In terms of product supply resilience, supplier diversification is at an excellent level, product quantity resilience and product supply efficiency are at an average level, and inventory management is at a good level. In terms of economic resilience, its price resilience is at an average level, and other indicators are at a good level. In terms of logistics resilience, its storage resilience is at an excellent level, and other indicators are at a good level. In terms of risk management and recovery ability, the risk identification and assessment of the supply chain are at a good level, and the risk management strategy and recovery speed and effectiveness are at an excellent level.

C (E-commerce-led forest products platform supply chain): In terms of production resilience, its raw material availability, process resilience, and partner resilience are at an excellent level, but its technological innovation ability is at an average level and other indicators are at a good level. In terms of product supply resilience, supplier diversification, product quantity resilience, and product supply efficiency are at an excellent level, and inventory management is at a good level. In terms of economic resilience, the profit growth rate, foreign trade dependence, and innovation output is at an excellent level, and the other indicators are at a good level. In terms of logistics resilience, in addition to the logistics network at a good level, other indicators are in the excellent level. In terms of risk management and resilience, the supply chain’s risk identification and assessment, risk management strategy, and recovery speed and effectiveness are all at an excellent level.

The weight value of each indicator and the correlation degree of each indicator are substituted into Formula (7). Based on this, the comprehensive correlation degree of the three first-level indicators of supply chain resilience evaluation A, B, and C is obtained, respectively, so as to evaluate the correlation degree of the first-level indicators of supply chain resilience. The calculation results are shown in Table 6 and Table 7.

It is not difficult to see from Table 6 and Table 7 that among the five levels of resilience evaluation of the forest products platform supply chain, A is at an excellent level in terms of production resilience, product supply resilience, logistics resilience, and risk management and resilience, while only economic resilience is at a good level. B is only at an excellent level in risk management and recovery ability, while the other four levels are at a good level. C, like A, is only at a good level of economic resilience, and all other levels are at an excellent level.

In summary, the resilience level of the three supply chains can be obtained, as shown in Table 8.

From the evaluation results, we can intuitively see the resilience level of the three main dominant models of the forest products platform supply chain. The resilience level of A (Integrated forest products platform supply chain) is “excellent”, B (Supplier-led forest products platform supply chain) is “good”, and C (E-commerce-led forest products platform supply chain) is “excellent”. The evaluation results are consistent with the actual situation.

The integrated forest products platform supply chain is more resilient than the supplier-led forest products platform supply chain. The integrated forest products platform supply chain has more efficient decision-making power in the face of emergencies, and it can quickly obtain market information in a short time. It can also provide a rapid response to market demand and a concentration of materials and ensure that the quality and quantity of forest products are delivered to the hands of consumers, to ensure the supply of forest products and stabilize prices.

The performance of the supplier-led forest products platform supply chain in production resilience, product supply resilience, economic resilience, and logistics resilience needs to be improved. The performance levels of these four dimensions are all good, so there is a large room for improvement. The reason why we attach importance to the improvement of these four dimensions is that these four dimensions have a profound impact on the operation of the entire supply chain. These aspects account for a large proportion of the supply chain resilience and play a key role in the improvement of supply chain resilience, so relevant performance targets should be set to improve them.

The advantage of the forest products platform supply chain led by the e-commerce platform is that it connects buyers and sellers, enhances the transparency of the market, and broadens the sales channels of forest products. E-commerce platforms provide suppliers with the opportunity to communicate directly with consumers and the ability to quickly adjust products and services based on customer feedback. At the same time, e-commerce platforms have promoted technological advances in the forest products supply chain, such as the application of technologies such as the Internet of Things (IoT) and blockchain, improving the supply chain’s ability to trace and control forest products. Finally, by reducing intermediate links, the supply chain can reduce transaction costs, including transportation, warehousing, and labor costs, thereby increasing the resilience of the supply chain.

5.4. Comparative Analysis-DEA Method

DEA is a cross-research method of operations research, management science, and mathematical economics that is used to evaluate the relative effectiveness of comparable units of the same type and has been widely used in the performance evaluation of computing, production, innovation, resource allocation, and other activities. Compared with traditional efficiency evaluation methods, DEA has the advantage in that it can handle multi-input and multi-output situations and automatically assign optimal weights to them, avoiding the influence of subjective factors, which makes the evaluation process error smaller and more objective, so as to carry out a more comprehensive and accurate efficiency evaluation.

The BCC model is used to evaluate pure technical efficiency. Its basic theory is to assume that the return to scale of the decision-making unit is variable, and evaluate after eliminating the scale-efficiency factor, so as to judge the proportional relationship between input and output. Therefore, this paper uses DEAP2.1 software and the DEA-BCC model to calculate the comprehensive technical efficiency, pure technical efficiency, scale efficiency, and scale return of the supply chain of three forest product platforms. The specific results are shown in Table 9. Comprehensive technical efficiency refers to the comprehensive measurement and evaluation of enterprise resource allocation ability, resource use efficiency, and other factors. In general, the comprehensive technical efficiency is equal to the product of pure technical efficiency and scale efficiency. If the comprehensive technical efficiency is equal to 1, that is, the combined efficiency of technical efficiency and scale efficiency is 1, it indicates that the comprehensive technical efficiency of this decision-making unit is effective.

It can be seen from Table 9 that the pure technical efficiency of the supply chain of the three forest product platforms is equal to 1, and the pure technical efficiency refers to the production efficiency of the enterprise affected by factors such as management and technical level, which is the production efficiency of the input factors of DMU when the scale efficiency is the best. Since the pure technical efficiency of A, B, and C are all 1, it indicates that under the current technology level, the production efficiency of the input factors is equal to 1. Its use of resources is efficient. Similarly, the scale efficiency of A, B, and C is also 1. Scale efficiency refers to the production efficiency affected by enterprise scale factors, which can truly reflect the difference between it and the optimal production level. Scale efficiency equals 1, indicating that the scale of the supply chain is in the best state with unchanged returns to scale, neither waste nor shortage of resource input, and the scale of input and output is just right. Finally, the comprehensive technical efficiency of A, B, and C is 1, which indicates that the overall performance of the three supply chains is good.

By comparing with the above research results, we find that the supply chain performance results calculated by the DEA method only provide a general reference, and its accuracy is not enough only to use this method for performance evaluation. In addition, although it is relatively simple to use this method and the way of data presentation is also intuitive, the use of the DEA method for performance evaluation cannot effectively help enterprises identify the problem. For each small indicator involved in performance evaluation, this method does not make specific explanations.

6. Discussion

From the evaluation results of first-level indicators specifically.

(1): Production resilience B₁

Both A (Integrated forest products platform supply chain) and C (E-commerce-led forest products platform supply chain) are at an excellent level in terms of production toughness, while only B (Supplier-led forest products platform supply chain) is at a good level, not excellent. Therefore, B needs to be further optimized in terms of production toughness. In terms of specific indicators, A and C are rated excellent in raw material availability, process resilience, and partner resilience, while B is rated good. For other indicators, A, B, and C all need to be further optimized and improved to achieve the best state.

(2): Product supply resilience B₂

A and C are at an excellent level in terms of product supply resilience, while B is at a good level in terms of product supply resilience. In terms of secondary indexes, A and C are in excellent grades in terms of supplier diversification, product quantity toughness, and product supply efficiency, which can ensure timely production and supply of forest products, but they can be further optimized in inventory management, and inventory management ability affects the improvement of product supply toughness to a certain extent. And B needs to focus on improving product quantity resilience and product supply efficiency, both of which are currently at an average level.

(3): Economic resilience B₃

The economic resilience levels of A, B, and C are all good, but not excellent. From the perspective of secondary indicators, the three supply chains need to improve in terms of market potential and price resilience. While A also needs to improve the foreign trade dependence and innovation output of the supply chain, B needs to improve the capital accumulation rate, profit growth rate, foreign trade dependence, and innovation output, and C only has shortcomings in the capital accumulation rate.

(4): Logistics resilience B₄

The level of logistics resilience of A and C is excellent, and B is good. As far as the secondary indicators are concerned, all the logistics indicators of A are excellent, and we should pay attention to maintaining this good state. C also needs to improve the flexibility of the logistics network, which is currently at a good level, while B needs to improve the overall logistics toughness from many aspects, such as transportation mode and information transmission.

(5): Risk management and resilience B₅

The resilience performance of different types of the forest products platform supply chain in this dimension is excellent. In terms of specific indicators, A is in a good state in terms of risk management strategy and needs to be further improved, while B is still at a good level in risk identification and assessment. Therefore, the important thing for B to pay attention to is the accurate prediction and assessment of risks in the supply chain operation process.

Based on the above analysis, the following suggestions were put forward to improve the resilience of the forest products platform supply chain:

(1): Implement strict risk management, which helps the supply chain to identify potential hazards and prevent supply chain disruptions by developing contingency plans. Secondly, strengthen the communication and exchange between the main bodies of the supply chain, strengthen the information construction of the platform supply chain, use blockchain and Internet of Things technology to enhance the transparency of the supply chain information of the forest products platform, establish a firm partnership, and promote information sharing, so as to provide the possibility of positive interaction between the members of the supply chain at the same level and improve the response ability of the supply chain.
(2): Implement the forest protection system, emphasizing sustainable logging and green plant cultivation, and protecting the natural ecosystem, so as to promote the ecological protection and sustainable development of forests. At the same time, the development of forest product processing, especially deep processing, to achieve transformation and value-added, promote the adjustment of the forest product structure and improve the comprehensive benefits of forest products and market competitiveness.
(3): Forestry trade should improve the distribution environment of forest products, change the distribution model of forest products, establish a common distribution system and logistics system, strengthen the integration ability of online and offline channels, innovate the sales method of forest products, and stimulate the vitality of the forest products sales market. Support should be given to the construction of forest product logistics and transportation, and at the same time, support should be given to the established forest product base in the testing and demonstration of new varieties and new technologies.

7. Conclusions

The research on the supply chain resilience evaluation of the forest products platform has become one of the hot spots in the field of forestry economics. However, few previous studies have assessed the level of resilience in the forest products platform supply chain. In order to make up for this shortcoming, in view of the objective and effective advantages of artificial intelligence, and the substantial existence of climate and transportation risks faced by forestry, this paper selects different types of forest product supply chains as research objects. Artificial intelligence and extension theory were used to evaluate the supply chain resilience of the forest products platform. Firstly, the paper uses artificial intelligence to select three leading models of the forest products platform supply chain: A (Integrated forest products platform supply chain), B (Supplier-led forest products platform supply chain), and C (E-commerce-led forest products platform supply chain). Secondly, a resilience evaluation index system was created in the artificial intelligence system combining the existing resilience indicators. Finally, the extension evaluation method was used to evaluate the resilience of three main forest product platform supply chains under different risk preferences. The results show that the supply chain resilience level of A is “excellent”, the supply chain resilience level of B is “good”, and the supply chain resilience level of C is “excellent”. This research contributes to the protection of natural ecosystems, the sustainable use of forestry resources, and economic development.

7.1. Implications

Theoretical contribution: Firstly, we defined the connotation and essence of supply chain resilience, which refers to the ability of the supply chain to quickly make plans and maintain normal operations by solving problems in the face of uncertain risks. Secondly, the article has created a resilience evaluation index system from five dimensions: production resilience, product supply resilience, economic resilience, logistics resilience, and risk management and recovery capability, enriching the supply chain resilience evaluation index system. Finally, the application of artificial intelligence and extension theory in supply chain resilience evaluation clarifies the steps for using this method, enriching the theoretical methods in the field of supply chain.

Practical significance: It is very important to study the supply chain of forest products platform for forest protection since the operation of the supply chain of forest products platform will cause some damage to the natural ecological environment, including over-logging, habitat destruction, and environmental pollution, etc. Therefore, the resilience and adaptability of the supply chain of forest products platform can be improved by assessing the resilience level of the supply chain. It also enables relevant enterprises to focus on environmental benefits as well as economic benefits, which is essential to ensure the sustainable development of the forestry economy and protect fragile natural ecosystems. At the same time, studying the supply chain resilience of the forest products platform can provide insight into specific forest-conservation measures and identify areas that need to be improved to protect forest resources. Finally, the results of the resilience assessment of the supply chain of the forest products platform can provide references for the formulation of government policies and regulations, which lays a foundation for the implementation of a series of strategies and actions to protect natural forests.

7.2. Limitations and Future Prospects

There are still some limitations and deficiencies. First, this study is limited by conditions. The sample size obtained in the analysis part is relatively limited. Second, the technology integration problem, the model construction is complex, the effective integration of AI recommendation system algorithm and extension system analysis tools need to overcome many obstacles, such as high technology purchase costs, data transmission problems, etc. Specifically, developing efficient artificial intelligence algorithms requires a significant amount of research work, which involves not only theoretical research but also multiple stages such as experimental verification. The entire process is time-consuming, leading to an increase in technology development costs. Secondly, extensible comprehensive evaluation requires the collection of a large amount of data, and obtaining this data often requires a significant investment of time and money, which increases the cost of data acquisition and processing. Finally, in order to maintain the operation of the expert team and its development, enterprises can only increase their personnel costs. Third, there is a gap between theoretical research and practical application. How to translate research results into implementable strategies and methods to solve the problems existing in the supply chain in real life still needs to be explored more in the future.

In summary, under the influence of different risk preferences, this paper evaluates the resilience of the supply chain of the forest products platform by using artificial intelligence and extension theory, solves the problems of index identification and method in the resilience assessment of the supply chain of the forest products platform, and optimizes the operational process of the supply chain of the forest products platform. The rationality of the application of artificial intelligence and extension theory in the evaluation of supply chain resilience is verified. In addition, the in-depth application of artificial intelligence and extension comprehensive evaluation method in the supply chain of forest products platform proves that the method can be extended to a wider range of research fields and provides innovative theoretical and methodological support for the study of supply chain resilience evaluation.

Author Contributions

Conceptualization, X.L. and P.L.; Methodology, X.L.; Formal analysis, L.L.; Investigation, P.L. and L.L.; Writing—original draft, P.L. and L.L.; Writing—review and editing, X.L. and P.L.; Funding acquisition, X.L. All authors have read and agreed to the published version of the manuscript.

Funding

This work was supported by Guangxi Key R&D Plan (Project No: 2022AB34029): The Key Technologies and Industrialization of Intelligent Traceability in the Whole Link of High-quality Seedling Seed Supply Chain in Lijiang River Basin and by Innovation Project of Guangxi Graduate Education (Project No: JGY2024045) and by the Research Fund Project of Development Institute of Zhujiang-Xijiang Economic Zone, Key Research Base of Humanities and Social Sciences in Guangxi Universities (Project No: ZX2023051).

Data Availability Statement

The corresponding author has the right to disclose the data upon reasonable request.

Conflicts of Interest

The authors declare no conflicts of interest.

References

Soumen, B.; Biswajeet, P.; Sambhunath, R.; Sengupta, D.; Bhunia, G.S.; Shit, P.K. Estimating Forest-Based Livelihood Strategies Focused on Accessibility of Market Demand and Forest Proximity. Small-Scale For. 2023, 22, 537–556. [Google Scholar]
Ahmad, M.M.; David, C.; James, R.; Narula, S.A. Climate change impacts on non-timber forest products: NTFP-dependent community responses from India. Clim. Dev. 2023, 15, 738–751. [Google Scholar]
Christian, T.; Harald, B.; Ché, E. Adaptive management for competing forest goods and services under climate change. Ecol. Appl. A Publ. Ecol. Soc. Am. 2012, 22, 65–77. [Google Scholar]
Björk, A.; Erlandsson, M.; Häkli, J.; Jaakkola, K.; Nilsson, Å.; Nummila, K.; Puntanen, V.; Sirkka, A. Monitoring environmental performance of the forestry supply chain using RFID. Comput. Ind. 2011, 62, 830–841. [Google Scholar] [CrossRef]
Cherodian, R.; Fraser, I. An environmental Kuznets curve for global forests: An application of the mi-lasso estimator. For. Policy Econ. 2024, 168, 103304. [Google Scholar] [CrossRef]
Teymourifar, A.; Trindade, M.A.M. A Framework to Design and Evaluate Green Contract Mechanisms for Forestry Supply Chains. Sustainability 2023, 15, 7668. [Google Scholar] [CrossRef]
Childerhouse, P.; Aqqad, A.M.; Zhou, Q.; Bezuidenhout, C. Network resilience modelling: A New Zealand forestry supply chain case. Int. J. Logist. Manag. 2020, 31, 291–311. [Google Scholar] [CrossRef]
Tavana, M.; Kaviani, A.M.; Caprio, D.D.; Rahpeyma, B. A two-stage data envelopment analysis model for measuring performance in three-level supply chains. Measurement 2016, 78, 322–333. [Google Scholar] [CrossRef]
Ribeiro, P.J.; Barbosa-Povoa, A. Supply Chain Resilience: Definitions and quantitative modelling approaches—A literature review. Comput. Ind. Eng. 2018, 115, 109–120. [Google Scholar] [CrossRef]
Guo, L.; Jing, X.; Na, L.; Dmitry, I. Blockchain-supported business model design, supply chain resilience, and firm performance. Transp. Res. Part E 2022, 163, 102773. [Google Scholar]
Abubakar, A.; Amr, M.; Amr, A. Analysing supply chain resilience: Integrating the constructs in a concept mapping framework via a systematic literature review. Supply Chain. Manag. Int. J. 2017, 22, 16–39. [Google Scholar]
Kaisa, A.K.; Pietro, E.; Jukka, H.; Immonen, M.; Lintukangas, K. COVID-19 as a trigger for dynamic capability development and supply chain resilience improvement. Int. J. Prod. Res. 2023, 61, 2696–2715. [Google Scholar]
Kamalahmadi, M.; Mellat-Parast, M. Developing a resilient supply chain through supplier flexibility and reliability assessment. Int. J. Prod. Res. 2016, 54, 302–321. [Google Scholar] [CrossRef]
Wang, S.Y.; Tian, X.J. Research on Sustainable Closed-Loop Supply Chain Synergy in Forest Industry Based on High-Quality Development: A Case Study in Northeast China. Forests 2022, 13, 1587. [Google Scholar] [CrossRef]
Sembiring, N.; Napitupulu, H.L.; Sembiring, M.T.; Ishak, A.; Irwany, F. A review: Hybrid simulation in forestry supply chain. IOP Conf. Ser. Earth Environ. Sci. 2021, 912, 012009. [Google Scholar] [CrossRef]
Borges, F.A.; Laurindo, J.F.; Spínola, M.M.; Gonçalves, R.F.; Mattos, C.A. The strategic use of artificial intelligence in the digital era: Systematic literature review and future research directions. Int. J. Inf. Manag. 2020, 57, 102–225. [Google Scholar] [CrossRef]
Golan, E.; Baoselie. Traceability in the US food supply;dead end or superhighway. Choices 2014, 32, 5–9. [Google Scholar]
Jon, H. Contractibility and the Design of Research Agreements. Am. Econ. Rev. 2010, 100, 214–246. [Google Scholar]
Taylor, T.A. Supply chain coordination under channel rebates with sales effort effect. Manag. Sci. 2013, 48, 992–1007. [Google Scholar] [CrossRef]
Mighell, L.; Jones, L.A. Vertical Coordination in Agriculture Department of Agriculture Economic Research Service. Agric. Econ. Rep. 1963, 62, 2–9. [Google Scholar]
Panda, S. Pricing and Replenishment Policies in Dual-channel Supply Chain under Continuous unit Cost Decrease. Appl. Math. Comput. 2015, 256, 913–929. [Google Scholar] [CrossRef]
Rauch, P.; Gronalt, M. Evaluating organisational designs in the forestry wood supply chain to support Forest Owners Cooperations. Small-Scale For. 2007, 4, 53–68. [Google Scholar] [CrossRef]
Petrescu, D.; Petcou, C.; Baibarac, C. Co-producing commons-based resilience: Lessons from R-Urban. Build. Res. Inf. 2016, 44, 717–736. [Google Scholar] [CrossRef]
Li, J.; Fu, G.; Zhao, X. Urban Economic Resilience and Supply Chain Dynamics: Evaluating Monetary Recovery Policies in Global Cities during the Early COVID-19 Pandemic. Mathematics 2024, 12, 673. [Google Scholar] [CrossRef]
Robert, O.; Melo, L.A.; Godslove, A.; Tam, V. Systematic review of critical infrastructure resilience indicators. Constr. Innov. 2023, 23, 1210–1231. [Google Scholar]
Gardas, B.B.; Raut, R.D.; Narkhede, B. Evaluating critical causal factors for post-harvest losses (PHL) in the fruit and vegetables supply chain in India using the DEMATEL approach. J. Clean. Prod. 2018, 199, 47–61. [Google Scholar] [CrossRef]
Sharma, R.; Kannan, D.; Darbari, D.J.; Jha, P.C. Group decision making model for selection of performance indicators for sustainable supplier evaluation in agro-food supply chain. Int. J. Prod. Econ. 2024, 277, 109353. [Google Scholar] [CrossRef]
Yang, J. Construction of urban livability evaluation index system by principal component analysis combined with entropy value method. Appl. Math. Nonlinear Sci. 2024, 9, 1–9. [Google Scholar] [CrossRef]
Zhai, Y. Research on the construction of digital village evaluation model based on AHP-entropy TOPSIS method. Appl. Math. Nonlinear Sci. 2024, 9, 1–9. [Google Scholar] [CrossRef]
Geng, S.; Liu, S. An agent-based framework for resilience analysis of service networks. Reliab. Eng. Syst. Saf. 2025, 253, 110523. [Google Scholar] [CrossRef]
Liu, R. Resilience Assessment of Deep Excavation Engineering Systems Based on Bayesian Network. J. Innov. Dev. 2024, 8, 5–9. [Google Scholar] [CrossRef]
Martin, R.; Gardiner, B. The Resilience of Cities to Economic Shocks: A Tale of Four Recessions (and the Challenge of Brexit). Pap. Reg. Sci. 2019, 98, 1801–1832. [Google Scholar] [CrossRef]
Wang, H. Linking AI supply chain strength to sustainable development and innovation: A country-level analysis. Expert Syst. 2022, 41, e12973. [Google Scholar] [CrossRef]
Modgil, S.; Singh, R.K.; Hannibal, C. Artificial intelligence for supply chain resilience: Learning from COVID-19. Int. J. Logist. Manag. 2022, 33, 1246–1268. [Google Scholar] [CrossRef]
Ibnouf, M.; Jaber, H.; Abukhalifeh, H.; Ghazal, M.; Ramadan, M.; Alkhedher, M. A Comprehensive Review of AI Algorithms for Performance Prediction, Optimization, and Process Control in Desalination Systems. Desalination Water Treat. 2024, 321, 100892. [Google Scholar] [CrossRef]
Lu, J.; Wu, D.S.; Mao, M.S.; Zhang, G. Recommender system application developments: A survey. Decis. Support Syst. 2015, 74, 12–32. [Google Scholar] [CrossRef]
Zhang, Q.; Lu, J.; Jin, Y.C. Artificial intelligence in recommender systems. Complex Intell. Syst. 2021, 7, 439–457. [Google Scholar] [CrossRef]
Kollia, I.; Stevenson, J.; Kollias, S. AI-Enabled Efficient and Safe Food Supply Chain. Electronics 2021, 10, 1223. [Google Scholar] [CrossRef]
Jia, D.C.; Pei, Y.Z. Problems and Countermeasures of Forestry Technology Popularization in Ecological Forestry Construction. Agric. Technol. Equip. 2020, 3, 1–3. [Google Scholar]
Puntoni, S. Artificial Intelligence in Marketing: From Computer Science to Social Science. J. Macromarketing 2024, 44, 883–885. [Google Scholar] [CrossRef]
Liu, X.; Wang, Z.; Pan, D.; Zhang, Y.; Zhang, J.; Chen, Y. Evaluation of Operational Safety Risk in Wastewater Treatment Plants Based on WSR and Matter–Element Extension Theory. Water 2024, 16, 2925. [Google Scholar] [CrossRef]
Xu, X.; Zhao, Y.; Nyberg, T.; Bao, S. Study on optimisation of the food cold chain transportation service network based on the extenics. Int. J. Wirel. Mob. Comput. 2016, 11, 131–136. [Google Scholar] [CrossRef]
Yang, C.Y.; Cai, W. Proceedings of the 2nd International Symposium on Extenics and Innovation Methods. Taylor Fr. 2015, 1, 1–15. [Google Scholar]
Li, M.; Xu, K.; Huang, S. Evaluation of green and sustainable building project based on extension matter-element theory in smart city application. Comput. Intell. 2020, 40, e12286. [Google Scholar] [CrossRef]
Ng, D.K.W.; Cai, W. Treating non-compatibility problem from matter element analysis to extenics. ACM Sigice Bull. 1997, 22, 2–9. [Google Scholar] [CrossRef]
Li, S.Q.; Li, R.R. Energy Sustainability Evaluation Model Based on the Matter-Element Extension Method: A Case Study of Shandong Province, China. Sustainability 2017, 9, 2128. [Google Scholar] [CrossRef]
Shan, W.; Zhang, Q.P. Extension theory and its application in evaluation of independent innovation capability. Kybernetes 2009, 38, 457–467. [Google Scholar] [CrossRef]
Liu, H.; Liu, Y.L.; Nong, Y.; Liu, L. The Evaluation of Land Consolidation’s Benefits Based on Extensional Matter-Element Model; Wuhan University: Wuhan, China, 2008; Volume 7144, p. 71440. [Google Scholar]
Na, Y.Z.; Yan, L. Durability Evaluation of Concrete Structure Based on Fuzzy Extension AHP. Appl. Mech. Mater. 2013, 454, 179–182. [Google Scholar]
Zhang, Z.; Deng, Z. Extension AHP-Based Assessment of Safety Risk Levels in Subway Deep Excavation Construction. J. Civ. Eng. Urban Plan. 2023, 5, 48–60. [Google Scholar]

Figure 1. Forest products platform supply chain.

Table 1. Resilience Evaluation Index System and Data Sources.

Indicators	Secondary Indicators	Secondary Indicator Description	Data Source
Production Resilience (B₁)	Availability of raw materials (C₁)	The ability to access forest resources	Expert rating
	Process resilience (C₂)	Supply and marketing process smooth circulation degree	Expert rating
	Quality control system (C₃)	Strictly control the quality of forest products	Expert rating
	Technological innovation ability (C₄)	Innovation in production technology	Records of enterprise surveys
	Resilience culture (C₅)	A culture that enhances responsiveness	Expert rating
	Strategic resilience (C₆)	Enterprise strategy formulation in response to emergencies	Expert rating
	Partner resilience (C₇)	The number of collaborative partners that can be adjusted in time	Public database
	Resource recycling (C₈)	Recycling of forestry resources	Public database
Product supply resilience (B₂)	Supplier diversification (C₉)	The equilibrium of the number of suppliers in the supply chain	Public database
	Product quantity resilience (C₁₀)	The resilience of product supply quantity	Public database
	Product supply efficiency (C₁₁)	The ability to deliver products to customers in a timely manner	Expert rating
	Inventory management (C₁₂)	Supply chain inventory management	Enterprise information statistics
Economic resilience (B₃)	Capital accumulation rate (C₁₃)	Total supply chain capital increased year-on-year or quarter-on-quarter	Financial statements
	Profit growth rate (C₁₄)	Profit growth	Financial statements
	Market potential (C₁₅)	Evaluate the possibility of market share growth	Expert rating
	Foreign trade dependence (C₁₆)	Total imports and exports of goods/GDP	Enterprise information statistics
	Price resilience (C₁₇)	As demand changes	Enterprise information statistics
	Innovation output (C₁₈)	Technology market turnover	Enterprise information statistics
Logistics resilience (B₄)	Diversity of transport modes (C₁₉)	The proportion of different modes of transport available	Enterprise information statistics
	Transportation reliability (C₂₀)	Value of product lost/Total value of product shipped	Enterprise information statistics
	Storage resilience (C₂₁)	The capacity of supply chain warehouse in the face of uncertainty	Enterprise information statistics
	Distribution resilience (C₂₂)	The ability of the distribution network to respond to emergencies	Enterprise information statistics
	Information timeliness rate (C₂₃)	Total number of timely messages/total number of messages delivered	Public database
	Delivery timeliness (C₂₄)	Just-in-time deliveries/total deliveries	Enterprise information statistics
	Logistics network (C₂₅)	The rationality of network design	Expert rating
	Requirements response speed (C₂₆)	Ability to respond to customer needs in a timely manner	Enterprise information statistics
Risk management and resilience (B₅)	Risk identification and assessment (C₂₇)	Identify potential risks and periodically assess risk levels	Public database
	Risk management strategy (C₂₈)	Detailed risk response within the supply chain	Public database
	Recovery speed and effectiveness (C₂₉)	Speed and quality of return to normal operations after a crisis	Public database

Table 2. Expert Scores of Indicators.

Indicators	Secondary Indicators	A	B	C
Production resilience B₁	C₁	4.02	3.83	4.21
	C₂	4.28	3.47	4.23
	C₃	3.95	3.85	3.16
	C₄	3.52	2.93	2.52
	C₅	3.23	2.99	3.34
	C₆	3.90	3.46	3.57
	C₇	4.20	2.46	4.12
	C₈	3.45	3.13	3.64
Product supply resilience B₂	C₉	4.27	4.08	4.26
	C₁₀	4.28	2.93	4.04
	C₁₁	4.13	2.40	4.23
	C₁₂	3.79	3.85	3.81
Economic resilience B₃	C₁₃	4.01	3.45	3.69
	C₁₄	4.13	3.13	4.23
	C₁₅	3.25	3.78	3.4
	C₁₆	3.32	3.07	4.21
	C₁₇	3.02	2.86	3.73
	C₁₈	3.11	3.77	4.06
Logistics resilience B₄	C₁₉	4.22	3.67	4.19
	C₂₀	4.18	3.97	4.20
	C₂₁	4.09	4.01	4.1
	C₂₂	4.13	3.97	4.21
	C₂₃	4.18	3.74	4.08
	C₂₄	4.27	3.77	4.17
	C₂₅	4.18	3.67	3.03
	C₂₆	4.19	3.92	4.23
Risk management and resilience B₅	C₂₇	4.24	3.89	4.15
	C₂₈	3.90	4.06	4.31
	C₂₉	4.11	4.08	4.22

Table 3. Judgment Matrix and Weight Values.

A	B₁	B₂	B₃	B₄	B₅	Weight
B₁	1	2	3	2	3	0.3571
B₂	1/2	1	2	1/2	2	0.1760
B₃	1/3	1/2	1	1/2	1/2	0.0953
B₄	1/2	2	2	1	3	0.2535
B₅	1/3	1/2	2	1/3	1	0.1181
Note: λmax = 5.16 CR = 0.036 < 0.1 Pass consistency test
B₁	C₁	C₂	C₃	C₄	C₅	C₆	C₇	C₈	Weight
C₁	1	2	3	2	4	3	3	4	0.2626
C₂	1/2	1	2	2	3	2	3	3	0.1872
C₃	1/3	1/2	1	1/2	2	3	2	3	0.1234
C₄	1/2	1/2	2	1	3	2	3	3	0.1579
C₅	1/4	1/3	1/2	1/3	1	1/2	1/3	1/2	0.0461
C₆	1/3	1/2	1/3	1/2	2	1	1/2	2	0.0762
C₇	1/3	1/3	1/2	1/3	3	2	1	2	0.0914
C₈	1/4	1/3	1/3	1/3	2	1/2	1/2	1	0.0552
Note: λmax = 8.408 CR = 0.041 < 0.1 Pass consistency test
B₂	C₉	C₁₀	C₁₁	C₁₂	Weight
C₉	1	4	3	2	0.4658
C₁₀	1/4	1	1/2	1/3	0.0960
C₁₁	1/3	2	1	1/2	0.1611
C₁₂	1/2	3	2	1	0.2771
Note: λmax = 4.031 CR = 0.012 < 0.1 Pass consistency test
B₃	C₁₃	C₁₄	C₁₅	C₁₆	C₁₇	C₁₈	Weight
C₁₃	1	1/3	1/3	2	1/2	1/4	0.0782
C₁₄	3	1	1/2	3	2	1/2	0.1803
C₁₅	3	2	1	3	2	1/2	0.2252
C₁₆	1/2	1/3	1/3	1	1/3	1/4	0.0585
C₁₇	2	1/2	1/2	3	1	1/3	0.1258
C₁₈	4	2	2	4	3	1	0.3320
Note: λmax = 6.175 CR = 0.028 < 0.1 Pass consistency test
B₄	C₁₉	C₂₀	C₂₁	C₂₂	C₂₃	C₂₄	C₂₅	C₂₆	Weight
C₁₉	1	2	1/2	1/3	1/4	1/4	1/3	1/2	0.0505
C₂₀	1/2	1	1/3	1/4	1/4	1/5	1/3	1/3	0.0366
C₂₁	2	3	1	1/2	1/3	1/3	1/2	1/3	0.0737
C₂₂	3	4	2	1	1/2	1/3	2	1/2	0.1293
C₂₃	4	4	3	2	1	2	2	3	0.2456
C₂₄	4	5	3	3	1/2	1	2	3	0.2213
C₂₅	3	3	2	1/2	1/2	1/2	1	2	0.1251
C₂₆	2	3	3	2	1/3	1/3	1/2	1	0.1179
Note: λmax = 8.489 CR = 0.05 < 0.1 Pass consistency test
B₅	C₂₇		C₂₈	C₂₉			Weight
C₂₇	1		1/2	1/3			0.1638
C₂₈	2		1	1/2			0.2972
C₂₉	3		2	1			0.5390
Note: λmax = 3.009 CR = 0.009 < 0.1 Pass consistency test

Table 4. Indicators correlation degree and grade of supply chain resilience evaluation of A and B.

Indicators	A					Grade	B					Grade
Indicators	N₁	N₂	N₃	N₄	N₅	Grade	N₁	N₂	N₃	N₄	N₅	Grade
C₁	−0.7550	−0.6733	−0.5100	−0.0200	0.0208	Excellent	−0.7075	−0.6100	−0.4150	0.1700	−0.1269	Good
C₂	−0.8200	−0.7600	−0.6400	−0.2800	0.6364	Excellent	−0.6175	−0.4900	−0.2350	0.4434	−0.2573	Good
C₃	−0.7375	−0.6500	−0.4750	0.0500	−0.0455	Good	−0.7125	−0.6167	−0.4250	0.1500	−0.1154	Good
C₄	−0.6300	−0.5067	−0.2600	0.4800	−0.2449	Good	−0.4825	−0.3100	0.0350	−0.0327	−0.3408	Average
C₅	−0.5575	−0.4100	−0.1150	0.1494	−0.3031	Good	−0.4975	−0.3300	0.0050	−0.0050	−0.3344	Average
C₆	−0.7250	−0.6333	−0.4500	0.1000	−0.0833	Good	−0.6150	−0.4867	−0.2300	0.4259	−0.2596	Good
C₇	−0.8000	−0.7333	−0.6000	−0.2000	0.3333	Excellent	−0.3724	−0.1575	0.2300	−0.1800	−0.3850	Average
C₈	−0.6125	−0.4833	−0.2250	0.4091	−0.2619	Good	−0.5325	−0.3767	−0.0650	0.0747	−0.3175	Good
C₉	−0.8175	−0.7567	−0.6350	−0.2700	0.5870	Excellent	−0.7700	−0.6933	−0.5400	−0.0800	0.0952	Excellent
C₁₀	−0.8200	−0.7600	−0.6400	−0.2800	0.6364	Excellent	−0.4825	−0.3100	0.0350	−0.0327	−0.3408	Average
C₁₁	−0.7825	−0.7100	−0.5650	−0.1300	0.1757	Excellent	−0.3684	−0.1429	0.2000	−0.2000	−0.4000	Average
C₁₂	−0.6975	−0.5967	−0.3950	0.2100	−0.1479	Good	−0.7125	−0.6167	−0.4250	0.1500	−0.1154	Good
C₁₃	−0.7525	−0.6700	−0.5050	−0.0100	0.0102	Excellent	−0.6125	−0.4833	−0.2250	0.4091	−0.2619	Good
C₁₄	−0.7825	−0.7100	−0.5650	−0.1300	0.1757	Excellent	−0.5325	−0.3767	−0.0650	0.0747	−0.3175	Good
C₁₅	−0.5625	−0.4167	−0.1250	0.1667	−0.3000	Good	−0.6950	−0.5933	−0.3900	0.2200	−0.1528	Good
C₁₆	−0.5800	−0.4400	−0.1600	0.2353	−0.2881	Good	−0.5175	−0.3567	−0.0350	0.0376	−0.3252	Good
C₁₇	−0.5050	−0.3400	−0.0100	0.0102	−0.3311	Good	−0.4650	−0.2867	0.0700	−0.0614	−0.3476	Average
C₁₈	−0.5275	−0.3700	−0.0550	0.0618	−0.3201	Good	−0.6925	−0.5900	−0.3850	0.2300	−0.1575	Good
C₁₉	−0.8050	−0.7400	−0.6100	−0.2200	0.3929	Excellent	−0.6675	−0.5567	−0.3350	0.3300	−0.1988	Good
C₂₀	−0.7950	−0.7267	−0.5900	−0.1800	0.2812	Excellent	−0.7425	−0.6567	−0.4850	0.0300	−0.0283	Good
C₂₁	−0.7725	−0.6967	−0.5450	−0.0900	0.1098	Excellent	−0.7525	−0.6700	−0.5050	−0.0100	0.0102	Excellent
C₂₂	−0.7825	−0.7100	−0.5650	−0.1300	0.1757	Excellent	−0.7425	−0.6567	−0.4850	0.0300	−0.0283	Good
C₂₃	−0.7950	−0.7267	−0.5900	−0.1800	0.2812	Excellent	−0.6850	−0.5800	−0.3700	0.2600	−0.1711	Good
C₂₄	−0.8175	−0.7567	−0.6350	−0.2700	0.5870	Excellent	−0.6925	−0.5900	−0.3850	0.2300	−0.1575	Good
C₂₅	−0.7950	−0.7267	−0.5900	−0.1800	0.2812	Excellent	−0.6675	−0.5567	−0.3350	0.3300	−0.1988	Good
C₂₆	−0.7975	−0.7300	−0.5950	−0.1900	0.3065	Excellent	−0.7300	−0.6400	−0.4600	0.0800	−0.0690	Good
C₂₇	−0.8100	−0.7467	−0.6200	−0.2400	0.4615	Excellent	−0.7225	−0.6300	−0.4450	0.1100	−0.0902	Good
C₂₈	−0.7250	−0.6333	−0.4500	0.1000	−0.0833	Good	−0.7650	−0.6867	−0.5300	−0.0600	0.0682	Excellent
C₂₉	−0.7775	−0.7033	−0.5550	−0.1100	0.1410	Excellent	−0.7700	−0.6933	−0.5400	−0.0800	0.0952	Excellent

Table 5. Indicators correlation degree and grade of supply chain resilience evaluation of C.

Indicators	C					Grade
Indicators	N₁	N₂	N₃	N₄	N₅	Grade
C₁	−0.8025	−0.7367	−0.6050	−0.2100	0.3621	Excellent
C₂	−0.8075	−0.7433	−0.6150	−0.2300	0.4259	Excellent
C₃	−0.5400	−0.3867	−0.0800	0.0952	−0.3134	Good
C₄	−0.3800	−0.1733	0.2400	−0.1622	−0.3737	Average
C₅	−0.5850	−0.4467	−0.1700	0.2576	−0.2845	Good
C₆	−0.6425	−0.5233	−0.2850	0.4300	−0.2312	Good
C₇	−0.7800	−0.7067	−0.5600	−0.1200	0.1579	Excellent
C₈	−0.6600	−0.5467	−0.3200	0.3600	−0.2093	Good
C₉	−0.8150	−0.7533	−0.6300	−0.2600	0.5417	Excellent
C₁₀	−0.7600	−0.6800	−0.5200	−0.0400	0.0435	Excellent
C₁₁	−0.8075	−0.7433	−0.6150	−0.2300	0.4259	Excellent
C₁₂	−0.7025	−0.6033	−0.4050	0.1900	−0.1377	Good
C₁₃	−0.6725	−0.5633	−0.3450	0.3100	−0.1914	Good
C₁₄	−0.8075	−0.7433	−0.6150	−0.2300	0.4259	Excellent
C₁₅	−0.6000	−0.4667	−0.2000	0.3333	−0.2727	Good
C₁₆	−0.8025	−0.7367	−0.6050	−0.2100	0.3621	Excellent
C₁₇	−0.6825	−0.5767	−0.3650	0.2700	−0.1753	Good
C₁₈	−0.7650	−0.6867	−0.5300	−0.0600	0.0682	Excellent
C₁₉	−0.7975	−0.7300	−0.5950	−0.1900	0.3065	Excellent
C₂₀	−0.8000	−0.7333	−0.6000	−0.2000	0.3333	Excellent
C₂₁	−0.7750	−0.7000	−0.5500	−0.1000	0.1250	Excellent
C₂₂	−0.8025	−0.7367	−0.6050	−0.2100	0.3621	Excellent
C₂₃	−0.7700	−0.6933	−0.5400	−0.0800	0.0952	Excellent
C₂₄	−0.7925	−0.7233	−0.5850	−0.1700	0.2576	Excellent
C₂₅	−0.5075	−0.3433	−0.0150	0.0155	−0.3299	Good
C₂₆	−0.8075	−0.7433	−0.6150	−0.2300	0.4259	Excellent
C₂₇	−0.7875	−0.7167	−0.5750	−0.1500	0.2143	Excellent
C₂₈	−0.8275	−0.7700	−0.6550	−0.3100	0.8158	Excellent
C₂₉	−0.8050	−0.7400	−0.6100	−0.2200	0.3929	Excellent

Table 6. Primary indicators correlation degree and grade of supply chain resilience evaluation of A and B.

Indicators	A					Grade	B					Grade
Indicators	N₁	N₂	N₃	N₄	N₅	Grade	N₁	N₂	N₃	N₄	N₅	Grade
B₁	−0.7301	−0.6402	−0.4603	0.0431	0.0760	Excellent	−0.5987	−0.4644	−0.1998	0.1609	−0.2374	Good
B₂	−0.7788	−0.7051	−0.5577	−0.1154	0.3218	Excellent	−0.6618	−0.5466	−0.3337	−0.0311	−0.0848	Good
B₃	−0.5992	−0.4656	−0.1984	0.0489	−0.1999	Good	−0.6191	−0.4921	−0.2382	0.1658	−0.2272	Good
B₄	−0.7975	−0.7300	−0.5950	−0.1900	0.3312	Excellent	−0.7034	−0.6045	−0.4068	0.1864	−0.1239	Good
B₅	−0.7672	−0.6896	−0.5344	−0.0689	0.1268	Excellent	−0.7607	−0.6810	−0.5215	−0.0429	0.0568	Excellent

Table 7. Primary indicators correlation degree and grade of supply chain resilience evaluation of C.

Indicators	C					Grade
Indicators	N₁	N₂	N₃	N₄	N₅	Grade
B₁	−0.6722	−0.5629	−0.3444	−0.0585	0.0493	Excellent
B₂	−0.7773	−0.7031	−0.5547	−0.1094	0.2869	Excellent
B₃	−0.7201	−0.6268	−0.4402	0.0596	0.0222	Good
B₄	−0.7536	−0.6715	−0.5072	−0.1339	0.1730	Excellent
B₅	−0.8088	−0.7451	−0.6176	−0.2353	0.4893	Excellent

Table 8. Grade of supply chain resilience evaluation of A, B, C.

	N₁	N₂	N₃	N₄	N₅	Grade
A	−0.7477	−0.6636	−0.4954	−0.0566	0.1637	N₅ Excellent
B	−0.6574	−0.5426	−0.3175	0.1100	−0.1460	N₄ Good
C	−0.7320	−0.6427	−0.4641	−0.0962	0.1719	N₅ Excellent

Table 9. Integrated technical efficiency of the platform supply chain of forest products.

Supply Chain	Technical Efficiency	Scale Efficiency	Overall Efficiency	Effectiveness
A	1	1	1	DEA strong and efficient
B	1	1	1	DEA strong and efficient
C	1	1	1	DEA strong and efficient

Disclaimer/Publisher’s Note: The statements, opinions and data contained in all publications are solely those of the individual author(s) and contributor(s) and not of MDPI and/or the editor(s). MDPI and/or the editor(s) disclaim responsibility for any injury to people or property resulting from any ideas, methods, instructions or products referred to in the content.

© 2024 by the authors. Licensee MDPI, Basel, Switzerland. This article is an open access article distributed under the terms and conditions of the Creative Commons Attribution (CC BY) license (https://creativecommons.org/licenses/by/4.0/).

Share and Cite

MDPI and ACS Style

Lu, L.; Long, P.; Luo, X. Resilience Evaluation of the Forest Products Platform Supply Chain Based on Artificial Intelligence and Extension Theory. Forests 2024, 15, 2180. https://doi.org/10.3390/f15122180

AMA Style

Lu L, Long P, Luo X. Resilience Evaluation of the Forest Products Platform Supply Chain Based on Artificial Intelligence and Extension Theory. Forests. 2024; 15(12):2180. https://doi.org/10.3390/f15122180

Chicago/Turabian Style

Lu, Lin, Ping Long, and Xiaochun Luo. 2024. "Resilience Evaluation of the Forest Products Platform Supply Chain Based on Artificial Intelligence and Extension Theory" Forests 15, no. 12: 2180. https://doi.org/10.3390/f15122180

APA Style

Lu, L., Long, P., & Luo, X. (2024). Resilience Evaluation of the Forest Products Platform Supply Chain Based on Artificial Intelligence and Extension Theory. Forests, 15(12), 2180. https://doi.org/10.3390/f15122180

Note that from the first issue of 2016, this journal uses article numbers instead of page numbers. See further details here.

Article Menu