Research on Extension Design of Emergency Cold Chain Logistics from the Perspective of Carbon Constraints

Abstract

Extenics has unique advantages in solving contradictions by using formal models to explore the possibility of expanding things and the laws and methods of development and innovation. This paper studies the specific application of the extension strategy generation method in emergency cold chain logistics, in order to solve the problem that the emergency plan is difficult to cover in the face of an emergency. The purpose of this paper is to provide ideas for the generation of strategies to solve the contradictions of cold chain logistics in complex emergency scenarios. Giving full play to the unique advantages of extenics in solving contradictory problems, this paper analyzes the core problems, objectives and conditions of emergency cold chain logistics in four links with the case scenario of the COVID-19 pandemic outbreak, extends and generates 10 measures to form 36 schemes, and evaluates the combination schemes quantitatively and objectively using the dependent function and superiority evaluation formula. In addition, the consideration of carbon constraints is added to the selection of the scheme, and the specific plan of integrating e-commerce platform, expert guidance, establishing temporary cold storage transfer and contactless distribution is designed. The research results provide support for meeting the needs of emergency logistics schemes in different situations and optimizing the energy efficiency of the scheme while ensuring humanitarian support. At the same time, the application of extenics basic-element formal language also provides a reference for further applying artificial intelligence to the design of emergency logistics schemes.

1. Introduction

Droughts, earthquakes, floods, storms, epidemics, nuclear explosions or chemical explosions [1], the frequent occurrences of disasters have brought serious challenges to the sustainable development of society [2]. If we cannot respond in time, it may also cause secondary injuries such as food shortages, panic riots and plague, which seriously threaten the safety of local people’s lives and property and affect regional economic development. In response to various large-scale disasters, many countries in the world have established emergency response mechanisms to achieve the purpose of humanitarian relief and accelerate the recovery of local economic production activities through scientific and efficient pre-disaster early warning, on-site rescue and post recovery emergency plans.

However, the type and intensity of disasters vary, and the timing and environment are very complex. Unconventional disasters often evolve [3] and produce chain reactions [4]. Emergency plans developed by relying on past cases and experiences will be much less practical in the face of some disasters that are beyond expectation or unprecedented, such as the 2019 Australian mountain fires, the 2021 Louisiana hurricane in the United States, the 2021 extraordinarily heavy rainfall in Henan Province, China, and the current sweeping global COVID-19 pandemic. They all seriously threaten the safety of life and property of local people and cause great losses to local economic development. In the face of such sudden disasters, how to quickly launch actions to protect the basic survival needs of local people and lay the foundation for restoring normal economic and social operations in the future has become the primary problem of emergency planning. In case of an emergency, a reliable emergency logistics system is vital for maintaining social stability [5].

Emergency cold chain logistics is an important channel to ensure food supply for people in disaster areas; however, the transportation conditions of cold chain logistics are difficult to adapt to the complex disaster environment. In order to enhance the scope of application of emergency plans to improve the rescue efficiency, and avoid the cost rise caused by excessive prevention, it is necessary to find a theoretical tool that can quickly generate response strategies to ensure the safety of people’s lives and property in the event of an emergency without increasing the cost burden of the emergency plan.

Taking contradictory problems as the research object, extenics discusses the laws and methods to solve them by formalization [6]. Its basic theory is extension theory, and its method system is extension innovation method. The application of extension theory and the extension innovation method in various fields is called extension engineering [7]. Extension theory takes basic-element as cells to formally describe things and their relationships in the world, establishes extension models as formal tools to solve contradictory problems, and represents the process of dealing with contradictory problems [7]. The extension innovation method is a formal and quantitative method used to expand, transform and evaluate the research object to generate ideas to solve various contradictory problems [8]. Since extenics has unique advantages in innovative design and outstanding problem-solving ability in the face of unfamiliar emergency situations, it is very suitable as a solution design tool for emergency cold chain logistics, which ca not only fill the gap of some emergency logistics rescue plans, but also provide reference for the design of other emergency plans in the future.

The purpose of this paper is to provide a specific solution strategy generation method combining qualitative and quantitative methods for cold chain logistics contradictions in complex emergency scenarios so as to meet the needs of humanitarian logistics rescue. In this paper, we contribute in two aspects. Firstly, we introduce the logic of extension logic and analyze the contradictory problems in emergency cold chain logistics by means of the formal language of basic-element. Compared with other analysis methods, the multidimensional triad of basic-element has the characteristics of concise content and clear expression, and at the same time, because its formal language is more easily recognized and learned by computers, it is more conducive to = future artificial intelligence to provide assistance in the design of emergency logistics solutions. Secondly, this paper evaluates the applicability of the proposed alternatives by means of superiority evaluation, and provides more objective and reasonable emergency cold chain logistics solutions by avoiding the bias brought by human subjectivity to a large extent with the help of the extension set and dependent function calculation. On this basis, we evaluate the total carbon emissions of the solutions with a higher degree of superiority, which helps to achieve low-carbon emission reduction while ensuring the humanitarian relief of emergency logistics, and provides a theoretical basis for emergency logistics research to improve energy utilization and achieve low-carbon emission reduction. Of course, the emergency fresh food supply chain is only an application branch, which cannot represent the whole emergency cold chain logistics. In the case selected in this paper, only the fresh food cold chain problem with the most urgent and largest demand is analyzed and discussed, which is representative for proving the application of extension design in the emergency cold chain logistics and can provide theoretical support for the combination of extenics theory and humanitarian logistics practice in the future.

This paper is structured as follows. The second part is a summary of the literature. In the third part, we analyze the contradictory problems of emergency cold chain logistics from demand collection to terminal distribution under the long-term closure policy against the background of COVID-19 being found for the first time in China. In the fourth part we introduce the theory of primitive analysis of topology and the calculation method of the correlation link function of superiority evaluation. In the fifth part, we find the key conflicts of the contradictory problems by primitive analysis for the background environment, expand the countermeasures using expandable analysis methods such as divergence trees, and initially confirm the feasible countermeasures with the help of the correlation function formula. Then we use the superiority evaluation to identify several more suitable solutions and use carbon emissions as the basis for the final solution selection. The sixth part is the conclusion and discussion.

2. Literature Review

A systematic literature review (SLR) is a well-known research method [9], the PRISMA Statement can be used as a basis for reporting systematic reviews of research, particularly evaluations of interventions [10]. According to the PRISMA 2020 statement [11], this paper chooses Web of Science and China National Knowledge Infrastructure databases to retrieve relevant research published up to June 2022, excluding articles unrelated to supply chain management, using the keyword combination of humanitarianism, emergency management, emergency cold chain, COVID-19 and low carbon.

Anisya and Mitsuko [12] defined humanitarian logistics as “The process of planning, implementing and controlling the efficient, cost-effective flow and storage of goods and materials as well as related information, from the point of origin to the point of consumption for the purpose of meeting the end beneficiary’s requirements”. As a direct means to optimize logistics efficiency, the location of emergency logistics facilities has become the entry point for many studies. Through reasonable planning of the location of distribution centers and other facilities, the optimization of material transportation efficiency can be achieved, including pre-positioned warehouse location selection [13], reduce logistics operating cost [14], improve response time [15], and the trade-off between logistics cost and response time [16], etc. Boonmee divides the facility location optimization model into four categories: deterministic, stochastic, dynamic, and robust [17]. Tofighi tries to achieve a balance between important performance such as response time, demand satisfaction and cost efficiency [18]. Haghi has added the consideration of the maximum justly service delivery to the casualties in affected areas [19]. Sheu proposed prioritized grouping of affected areas and time-varying relief demand forecasting to achieve dynamic relief supply [20]. Some other scholars focus on the optimization of vehicle transportation networks, Li [21] and Özdamar [22] expect to plan logistics at the macro level and to be able to dynamically update the logistics plan. Yi establishes a temporary emergency unit so as to realize the optimization of the route [23]. Sheu adjusts the resource transportation strategy based on the priority and the dynamic supply of the disaster area [20].

Efficient relief logistics scheduling depends on the accurate transport time information for available routes [24]. Ahmadi obtains real-time information data through a geographic information system to solve the problem of transportation uncertainty [14]. Feizollahi proved that a good database to provide necessary information is particularly important for emergency logistics operations [25]. L’Hermitte proposed to establish a common data platform to promote the interaction and trust of rescue organizations [26]. Jiang found that most of the articles ignore the synergy required by emergency logistics for the various units of the supply chain system [27]. Cozzolino explored the challenges of emergency logistics using cross-sectoral cooperation in humanitarian relief supply chain networks as an entry point [28]. An efficient information interaction system will provide good support for the supply of emergency resources.

In the rescue activities, fresh food has high requirements for transportation equipment, environment, timeliness and distribution methods [29], and cold chain logistics is an important channel to ensure its supply. Zhang [30], Xiong [31], Yan [32] and Theophilus [33] use the improved algorithm, Chen [34] applied cloud computing and big data, to realize the path optimization of cold chain logistics. Liu proposed to establish joint distribution nodes to solve the problem of insufficient terminal logistics distribution capacity of regional distribution centers [35]. Ali verified the relationship between cold chain logistics risk, flexibility and enterprise performance by establishing a cold chain logistics flexibility model [36]. Yang proposes a multidimensional robust optimization model to optimize multiple distribution centers and the soft time window problem in the context of COVID-19 [37]. Qi and Hu minimize the costs of loss on the premise of the shortest resource scheduling time [38]. Li builds an evaluation and strategy model for emergency cold chain supply chain synergy based on supply chain synergy theory [39].

The emergence of COVID-19 has posed a more severe challenge to the global food supply logistics, the burden of it on food supply chains has clearly affected demand and caused a shortage of transportation [40]. In recent research, Kazancoglu said that information sharing and managerial approaches play an important role in achieving food supply chains’ sustainability [41]. Yui-yip’s findings suggest critical points on the design and implementation of emergency logistics operations during a chaotic period [42]. Cardoso’s review proposes a taxonomy of impacts on four supply chain links: demand-side, supply-side, logistics and infrastructure, and management and operation [43]. Khamis studied the role of robotics and automation in responding to the pandemic [44]. Wincewicz-Bosy supposed that the pandemic undoubtedly caused the need to revise the existing solutions used in the food supply chain (external and internal) in order to ensure the continuity of the feeding processes [45]. Takavakoglou studied the application of constructed wetlands in the food supply chain to meet the challenges of COVID-19 [46].

The specialization of teams participating in the operation are one of vital life-saving coordination roles [47]. Rogers [48] and Mallik [49] emphasized the importance of professional training for cold chain transportation and management personnel. Zimmer believed volunteers need training to integrate into disaster relief agencies [50]. Paton said that training can reduce the impact of disaster uncertainty [51]. Hye suggested subdividing different fields so that professional volunteers can participate in disaster management [52]. Ghodsi believes that disaster related training is an important factor affecting the ability of volunteers to resist disasters [53].

Liu reported that the incomplete infrastructures of the cold chain will be one of the main reasons for transportation loss [54]. However, in recent years, there is little research on the emergency solution to the problem of insufficient logistics facilities. Most of the research, as mentioned before, focuses on the strategic level of precaution or rescue facilities network planning, and the specific problem-solving measures at the implementation level are insufficient. Huang [55], Khare [56] and Noyan [57] are all committed to optimizing the distribution network of the last mile logistics with algorithms. Moshref-Javadi studies drone-based delivery systems [58], Fikar presents a simulation and optimization-based decision-support system to improve disaster relief distribution [59], and Ferrer builds a compromise programming model for multi-criteria optimization in humanitarian last mile distribution [60]. However, in emergency rescue activities, it is sometimes necessary to find a fast and feasible solution rather than an unrealistic complex optimal solution, because time is crucial to emergency response [27].

Extenics has unique advantages in strategy generation [61,62,63] and performance evaluation [64,65] to solve contradictory problems. Zhu established the emergency coordination model of supply chain emergencies according to the method of extenics [66]. Wang established the earthquake emergency plan generation model based on the idea of scenario balance [67]. Ren accelerated product low-carbon design based on the extenics problem solving model [68,69]. Applying extenics to the generation of strategies to solve the contradictory problem of emergency cold chain logistics is conducive to improving the speed of emergency response and achieving higher satisfaction with emergency logistics services.

In addition, the impact of carbon dioxide emissions on the global climate is becoming more and more obvious. In recent years, the hypothesis that climate change leads to an extreme climate has been strongly confirmed. Shen [70], Wang [71], Zhang [72], Zhang [73] and Stellingwerf [74] have added the discussion of carbon emission cost into the location and path optimization model of logistics network. Wang takes the overall cost of cold chain logistics as the objective function to explore the vehicle route optimization under the dual constraints of time window and carbon emissions [75]. Qin took the minimum cost per satisfied customer as the objective function, and proposed that joint distribution and transportation will become the road for the long-term development of logistics enterprises [76]. Liu said that the government encourages logistics enterprises to adopt joint distribution of resource sharing through reasonable carbon quota and carbon price, which will achieve the purpose of environmental and economic friendliness [77].

It can be seen that most scholars’ emergency logistics and cold chain logistics schemes are limited to single or a few problems such as facility location, network optimization and time window, and the overall response ability to sudden problems in complex environments is insufficient. However, the randomness and complexity of sudden disasters have extremely high requirements for the problem-solving ability of emergency logistics. When the available resources are inconsistent with the target conditions, most scholars choose to find another way to avoid obstacles rather than solve problems. In addition, there is little research on emergency cold chain logistics with environmental costs. Considering that the climate problem has become a major problem that needs to be solved urgently in the world, managers need to take low-carbon and environmental protection as one of the reference factors in the formulation of an emergency cold chain logistics plan.

Taking the transportation dilemma of fresh products in an urban community in the early stage of the COVID-19 outbreak in 2020 as an example, this paper analyzes the emergency cold chain logistics diploma from four aspects: information transmission, personnel training, transportation equipment selection, and the last mile distribution. Using the formal language of extension primitives describes the key elements of the conflict between objectives and conditions in the contradictory problems through the formal language of basic-element and uses the dependent function formula to digitally express the contradiction or compatibility relationship. Then, with the help of extensible analysis methods such as divergence tree and correlation network, the scheme elements are designed innovatively, and the superiority of the combined scheme is evaluated. Finally, the scheme with the lowest total carbon emission is selected among the several schemes with high superiority, and the low-carbon optimization of the scheme is realized on the basis of ensuring the high matching degree of the scheme design.

3. Background and Method

The COVID-19 pandemic broke out in many countries within half a year from December 2019. Given that effective medical countermeasures did not exist in early 2020, governments had to adopt non-medical measures [78], including case isolation, the closure of schools and universities, banning of mass gatherings, public events, and local and national lockdowns [79]. The Chinese government first adopted a strict lockdown policy to avoid the further spread of the pandemic [80]. European countries introduced non-pharmaceutical interventions similar to those used in China [81]. Italy was the first country in Europe to apply intervention measures from the beginning of March 2020. Other EU countries followed soon afterward, using similar countermeasures around mid-March 2020 [78]. The stay-at-home requirements of closure policy avoids the gathering of people, greatly inhibits the movement and contact of people, controls the spread speed of the pandemic [82,83], and provides a prerequisite for further control and elimination of the pandemic.

However, the closure policy of stay-at-home requirements has restricted residents’ shopping activities, resulting in the pressure of food supply concentrated on the government. To this end, China has adopted community-based lockdown policies to arrange volunteers to take charge of unified management and carry out the statistics and distribution of procurement needs [84], which not only solves the problem of residents’ fresh food distribution, but also effectively blocked the spread of the virus. This method does solve the important problem of food demand of urban residents [85], but there are also many deficiencies.

3.1. Insufficient Informatization

The development of emergency logistics in China started relatively late, the modern information management of emergency logistics is still at a low level [86], and the prevention and grid management of communities generally lack the support of informatization and big data means [87]. Especially during the COVID-19 pandemic, the relationship between supply and demand was impacted, and the response speed of the supply chain became slower [88], so the interaction of supply and demand information of residents’ cold chain food became a major problem.

Due to the lack of a public and unified emergency logistics information exchange platform [89], the types of fresh food actually supplied are relatively minimal [90], and the needs of residents are difficult to transmit to every fresh food supplier. Although many suppliers hold a wide range of fresh products, they are unable to obtain residents’ orders and provide products to them. It is difficult to connect the supply and demand information, thus many fresh products rotted in the warehouse, but residents could not buy the food they wanted.

On the other hand, due to insufficient informatization, although food safety testing is stricter during the pandemic, most of the food lacks information tracing means, resulting in residents’ lack of trust in the safety of the food they received. Once some food is detected to be contaminated, the recycling of the same batch of food is inconvenient and it is difficult to find the source of pollution. There is a contradiction between the need for supply and demand information transmission and the current situation of insufficient informatization.

3.2. Insufficient Professionalism

Emergency events require that emergency logistics must be pertinent and professional [78]. One of the main reasons for the food virus input of Hailian cold storage in Tianjin eco city and cold chain in Longgang District of Shenzhen is the non-standard operation of the logistics transportation process [91]. A large number of scholars attach importance to the specialization of emergency logistics’ participation and regard it as one of the factors to ensure the smooth operation of emergency logistics activities [47,48,49,50,51,52,53].

In many stay-at-home requirement areas, the transportation and distribution teams are composed of doctors and volunteers [92], who play the role of linking material, human and other resources [93], but lack the knowledge and experience of professional cold chain logistics transportation [94]. On the one hand, the overlapping or too long transportation route planning will lead to low transportation efficiency and will take a long time, which will not only affect the food supply of residents in stay-at-home requirements areas, but also cause a lot of unnecessary energy waste and carbon emission. On the other hand, the non-standard operation of transportation personnel in the transportation link may lead to the exposure of fresh food to the virus environment, and the transmission of virus contaminated food in any link in the supply chain, forming a chain transmission [95] will greatly threaten the safety of residents at the end of the supply chain and hinder the pandemic prevention process. The lack of professionalism in the cold chain transportation process has led to a significant reduction in the timeliness and safety of the transportation of fresh products, which is in contradiction with the residents’ demand for food transportation in the pandemic environment.

3.3. Lack of Storage and Distribution Infrastructure

Since temperature abuse in the cold chain can cause microbial hazards and losses of product quality, temperature control is essential to keep the final consumer safe [88]. Cold storage refrigerated trucks and other facilities are important tools to ensure the environmental temperature of materials in cold chain transportation, and the incomplete cold chain infrastructure will be one of the main reasons for transportation losses [54]. The global spread of COVID-19 has led to a serious shortage of experts, equipment and capabilities [96].

Taking China as an example, the total capacity of cold storage is 130 million cubic meters, and the per capita cold storage area is only about 0.1 cubic meters/person, which is far lower than the level of 0.4 cubic meters/person in developed countries [97], refrigerated trucks accounting for only 0.3% of freight cars [98]. The imperfect facilities lead to the long distribution path and high loss rate of fresh products, and the vehicle scheduling is difficult to sustain. In such incidents as the transportation of residents’ meat products by general vehicles, the “broken chain” of cold chain logistics leads to the reduction of food safety, and greatly reduces the sense of security and trust of residents in the closure.

3.4. Last-Mile Logistics Problem

Scholars’ research on material delivery focuses on the transportation link of the last kilometer of transportation, including the selection and optimization of a distribution network [58,59,60,61] and delivery tools [62], that is, how to reach the target location from the distribution center. How the receiver picks up the goods after the goods arrive appears not to be a problem in the past. However, in the context of the COVID-19 pandemic, cold chain delivery has a great potential of contributing to the spread of COVID-19 [99].

The last mile of fresh food transportation under the pandemic is a challenge. Under the stay-at-home requirements and the home isolation policy, the final link of cold chain logistics is generally achieved by the mode of volunteers delivering goods to their homes or residents picking up goods at the designated place [100]. The presence of potential infected individuals increases the COVID-19 transmission risk during the delivery, both of the delivery modes have the risk of infection as the deliveryman may be the medium through which the virus spreads among urban residents [101] in the former mode, and cross-contamination may result from residents’ contact with each other in the latter mode. It remains a challenge to avoid human contact and to ensure that fresh food reaches residents in a safe and timely manner.

3.5. Method

Extenics takes basic-element as the logical cells for analyzing things. The movement of things and the relationship between things correspond to three primitives: matter-element, affair-element and relation-element, respectively. Taking one-dimensional matter-element as an example, the $O_m$ is the object, $c_m$ is the feature, the value of $O_m$ about $c_m$ is $v_m$, the three form an ordered triple:

$$\mathrm{M}=\left(\begin{array}{ccc}{O}_m,& c_m,& v_m\end{array}\right)$$

(1)

Since objects generally have multiple features, the multi-dimensional expression of basic-elements is helpful to express the diversified nature and movement relationship of things in detail [102]. The $O_m$ is the object, which has $n$ features $c_{m1}$, $c_{m2}$, …, $c_{mn}$, the value of about $c_{mi}$ ($i$ = 1, 2, …, $n$) is $v_{mi}$ ( = 1, 2, …, $n$), the multidimensional basic-element can be expressed as follows:

$$M=\left(\begin{array}{ccc}{O}_m,& C_m,& V_m\end{array}\right)=\left[\begin{array}{ccc}{O}_m,& c_{m1},& v_{m1}\\ & c_{m2},& v_{m2}\\ & ⋮& ⋮\\ & c_{mn},& v_{mn}\end{array}\right]$$

(2)

To solve the contradictory problem, first of all, it is necessary to clearly define the targets $G$ and conditions $L$ of the problem, so the general expression of the contradiction problem is constructed:

$$P=G\ast L$$

(3)

Extract the objectives and conditions to obtain the key to the conflict of contradictory problems, that is, the target $g$ and condition $l$ of the core problem, and then establish the primitive model:

$$\begin{gathered} P=g\ast l \\ g=\left[\begin{array}{ccc}{O}_1,& c_1,& v_1\end{array}\right] \\ l=\left[\begin{array}{ccc}{O}_2,& c_2,& v_2\end{array}\right] \end{gathered}$$

(4)

where $V_1$ is generally the feasible region $\langle a,b\rangle$, which means that the needs of the target $g$ can be met within this range. Generally, there is a optimal value $M$ in the feasible region, which means that when the value of the object $O_2$ about the feature $c_2$ is $M$, the maximum satisfaction can be achieved; $v_2$ is generally a specific value $x$, indicating that the actual value of the object $O_2$ about the feature $c_2$ under the original conditions.

The problem $P=g\ast l$ that the target $g$ cannot be achieved under condition $l$ is called an incompatible problem, which is denoted as $g\uparrow l$. If, under condition $l$, target $g_1$ and $g_2$ cannot be achieved at the same time, the problem $P=\left(g_1\wedge g_2\right)\ast l$ is called the Antithetical problem and is recorded as $\left(g_1\wedge g_2\right)\uparrow l$.

When the targets and conditions of the core problem are defined, the dependent function $K$ of the contradictory problem can be established to objectively and quantitatively express the degree of compliance between the conditions and the objectives, providing a quantitative measurement tool for the subsequent evaluation and analysis of the generated strategies.

According to the continuity and the position of the optimal value $M$ in the feasible region, the form of dependent function formula $K$ is as follows:

When $x$ is continuous, $x\in \langle a,b\rangle$ and the optimal value $M\in \langle a, b\rangle$:

$$K=k\left(x\right)=\left\{\begin{array}{c}\frac{x-a}{M-a},x\le M\\ \frac{b-x}{b-M},x>M\end{array}\right.$$

(5)

In particular, when $M=\frac{\mathrm{a}+\mathrm{b}}{2}$:

$$K=k\left(x\right)=\left\{\begin{array}{c}\frac{2\left(x-a\right)}{b-a},x\le M\\ \frac{2\left(b-x\right)}{b-a},x>M\end{array}\right.$$

(6)

When $x$ is continuous, $x\in \langle a,b\rangle$ and the optimal value $M$ is equal to $a$ or $b$:

① When $M=a$:

$$K=k\left(x\right)=\left\{\begin{array}{c}\frac{x-a}{b-a},x<M\\ \frac{b-x}{b-a},x\ge M\end{array}\right.$$

(7)

② When $M=b$:

$$K=k\left(x\right)=\left\{\begin{array}{c}\frac{x-a}{b-a},x\le M\\ \frac{b-x}{b-a},x>M\end{array}\right.$$

(8)

When $x$ is discrete:

$$K=k\left(x\right)=\left\{\begin{array}{c}{A}_1(>0),x=a_1\\ A_2(>0),x=a_2\\ \dots ,\hfill \\ A_m(>0),x=a_m\\ 0,x=a_0\hfill \\ B_1(<0),x=b_1\\ B_2(<0),x=b_2\\ \dots ,\hfill \\ B_n(<0),x=b_n\end{array}\right.$$

(9)

The extension innovation methods such as divergence tree, correlation network and implication system are used for the target or conditional basic-element. The elements of the basic-element are transformed to obtain a large number of alternative basic-elements. When the correlation degree $k$ of an alternative basic-element is not negative, it means that the new target and the condition are no longer contradictory, so it can be used as the element of the solution. Now we introduce the divergence tree method and implication system method used in this paper.

Divergence tree is an extended analysis method based on divergence analysis, which is derived from the principles of “one object with multiple features” and “one feature with multiple objects” etc. For basic element $M=\left(\begin{array}{ccc}{O}_m,& c_m,& v_m\end{array}\right)$, by fixing the object $O_m$ in the basic element to find other features $c_m{}^{\prime }$ and the corresponding magnitude $v_m{}^{\prime }$, a new basic element $M^{\prime }=\left(\begin{array}{ccc}{O}_m,& c_m{}^{\prime },& {\mathrm{v}}_m{}^{\prime }\end{array}\right)$ can be obtained, or by fixing the feature $c_m$ in the basic element to find the object ${O_m}^{\prime }$ with the same feature and different value, a new primitive $M^{″}=\left(\begin{array}{ccc}{O_m}^{″},& c_m,& {\mathrm{v}}_m{}^{″}\end{array}\right)$ can be obtained. Finally, the extended basic element is put into the actual environment to combine multiple practical schemes.

Implication system is an extended analysis method based on implication analysis. For two basic-elements $B_1$ and $B_2$, if $B_1$ must be realized when $B_2$ is realized, it is called $B_2$ contains $B_1$. Therefore, we can find the lower basic =-element of the target or condition in the core problem. If the lower basic-element containing the original target and condition primitive no longer conflict, we can get the strategy to solve the problem. In this paper, a feasible scheme is usually formed by combining implication system with divergence tree.

Although all schemes with positive compatibility are feasible, there must be advantages and disadvantages between them. At this time, the superiority $G$ of the scheme is weighted according to the importance of the link. The weight coefficient ${\alpha }_i$ is given as a value in the interval $\left[0,1\right]$ according to the importance of the object, and ${{\displaystyle \sum }}_{i=1}^n{\alpha }_i=1$. The normalization compatibility $K_i{}^{\prime }$ is obtained after the compatibility is normalized, and the superiority evaluation formula of the overall scheme is constructed. A higher degree of superiority means that the solution is more suitable for solving the conflicting problem:

$$G={\displaystyle \sum }_{i=1}^n{\alpha }_i{K}_i{}^{\prime }$$

(10)

4. Results and Discussion

In order to solve the contradictions in the case, safely and effectively realize the distribution of fresh food for residents in the stay-at-home requirements area, maintain the supply of basic living materials for residents in the pandemic prevention area, and assist in the implementation of the pandemic prevention closure policy, we analyze the emergency cold chain logistics basic-element model in the case in four stages from MI₁ information interaction to MI₄ contactless delivery.

4.1. MI₁ Information Interaction

During the pandemic period, in order to avoid personal contact and block the spread of the pandemic, community residents completed the demand submission and payment for fresh products through a wechat group. Volunteers use the wechat group to inform the fresh food they can buy and the corresponding price, and inform all residents in the wechat group to submit the list of purchases through the “Solitaire” function, and transfer the account in advance. After the residents’ needs are submitted, volunteers make manual statistics to form a general list, and submit procurement needs to suppliers by phone or SMS. As an information statistics platform, wechat only has the function of demand collection. In order to clearly reflect the efficiency and visibility of information traceability in the fresh food supply chain, the information traceability efficiency is reflected from the values of very poor, poor, medium, good and excellent from 1 to 5 (the best is 5). The response time is the time from the uploading of demand information to the completion of upstream receiving statistics, and the optimal value is 0.5 h (considering that a too fast response speed will lead to error information that cannot be modified). The condition basic-element $L_1$, target primitive $G_1$ and discrete correlation degree $K_{12}$ of the core problem can be expressed as:

$$L_1=\left[\begin{array}{ccc}\mathrm{Wechat}\mathrm{platform}{M}_1,& \mathrm{Information}\mathrm{traceability},& 1\\ & \mathrm{Response}\mathrm{speed},& 5\mathrm{h}\end{array}\right]$$

(11)

$$G_1=\left[\begin{array}{ccc}\mathrm{Information}\mathrm{interaction},& \mathrm{Receiving}\mathrm{object},& \mathrm{Resident}\mathrm{demand}\\ & \mathrm{Information}\mathrm{traceability},& 〈3,5〉\\ & \mathrm{Response}\mathrm{speed},& 〈0,2〉\mathrm{h}\end{array}\right]$$

(12)

$$K_{12}=k\left(x\right)=\left\{\begin{array}{c}1,x=5\\ 0.5,x=4\\ 0,x=3\\ -0.5,x=2\\ -1,x=1\end{array}\right.$$

(13)

Implication analysis of the information interaction platform:

$$M_1\Leftarrow \left\{\begin{array}{c}{M}_{11}=\left(\begin{array}{ccc}{M}_{11},& \mathrm{Statistical}\mathrm{method},& \mathrm{Artificial}\end{array}\right)\hfill \\ M_{12}=\left(\begin{array}{ccc}{M}_{12},& \mathrm{Platform}\mathrm{Type},& \mathrm{Chat}\mathrm{Software}\end{array}\right)\hfill \\ M_{13}=\left(\begin{array}{ccc}{M}_{13},& \mathrm{Platform}\mathrm{Function},& \mathrm{Demand}\mathrm{Collection}\end{array}\right)\hfill \end{array}\right.$$

(14)

According to model (13) and model (5), it can be calculated that the correlation degree $K_{11}=-1$ and $K_{12}=-2$ of the information traceability and response time length of target $G_1$ and condition $L_1$ are negative, forming an incompatible problem. The problem model is recorded as $G_1\uparrow L_1$. Therefore, $a$ divergence analysis of $M_{11}$, $M_{12}$ was performed:

$$M_{11}⊣\left\{\begin{array}{c}{M}_{111}=\left(\begin{array}{ccc}{M}_{111},& \mathrm{Statistical}\mathrm{method},& \mathrm{None}\end{array}\right)\hfill \\ M_{112}=\left(\begin{array}{ccc}{M}_{122},& \mathrm{Statistical}\mathrm{method},& \mathrm{Semi}-\mathrm{intelligent}\end{array}\right)\hfill \\ M_{113}=\left(\begin{array}{ccc}{M}_{113},& \mathrm{Statistical}\mathrm{method},& \mathrm{Intelligent}\end{array}\right)\hfill \end{array}\right.$$

(15)

$$M_{12}⊣\left\{\begin{array}{c}{M}_{121}=\left(\begin{array}{ccc}{M}_{121},& \mathrm{Platform}\mathrm{Type},& \mathrm{Verbal}\mathrm{Communication}\end{array}\right)\hfill \\ M_{122}=\left(\begin{array}{ccc}{M}_{122},& \mathrm{Platform}\mathrm{Type},& \mathrm{E}-\mathrm{commerce}\mathrm{platform}\end{array}\right)\hfill \\ M_{123}=\left(\begin{array}{ccc}{M}_{121},& \mathrm{Platform}\mathrm{Type},& \mathrm{Newly}\mathrm{built}\mathrm{platform}\end{array}\right)\hfill \end{array}\right.$$

(16)

$$M_{13}⊣\left\{\begin{array}{c}{M}_{131}=\left(\begin{array}{ccc}{M}_{131},& \mathrm{Platform}\mathrm{Function},& \mathrm{Demand}\mathrm{Forecast}\end{array}\right)\hfill \\ M_{132}=\left(\begin{array}{ccc}{M}_{132},& \mathrm{Platform}\mathrm{Function},& \mathrm{Demand}\mathrm{Response}\end{array}\right)\hfill \\ M_{133}=\left(\begin{array}{ccc}{M}_{133},& \mathrm{Platform}\mathrm{Function},& \mathrm{Collection}\oplus \mathrm{Statistics}\oplus \mathrm{Upload}\end{array}\right)\hfill \end{array}\right.$$

(17)

Combined with the needs of residents for the function of the fresh food information interaction platform during the pandemic, the following divergence elements are selected as measure elements:

$$T{M}_1=\left\{\begin{array}{l}{M_1}^{\prime }=\left\{\begin{array}{l}{M}_{112}=\left(\begin{array}{ccc}{M}_{112},& \mathrm{Statistical}\mathrm{method},& \mathrm{Semi}-\mathrm{intelligent}\end{array}\right)\\ M_{122}=\left(\begin{array}{ccc}{M}_{122},& \mathrm{Platform}\mathrm{Type},& \mathrm{E}-\mathrm{commerce}\mathrm{platform}\end{array}\right)\\ M_{133}=\left(\begin{array}{ccc}{M}_{133},& \mathrm{Platform}\mathrm{Function},& \mathrm{Collection}\oplus \mathrm{Statistics}\oplus \mathrm{Upload}\end{array}\right)\end{array}\right.\\ {M_1}^{″}=\left\{\begin{array}{l}{M}_{113}=\left(\begin{array}{lll}{M}_{113},& \mathrm{Statistical}\mathrm{method},& \mathrm{Intelligent}\end{array}\right)\\ M_{123}=\left(\begin{array}{lll}{M}_{123},& \mathrm{Newly}\mathrm{built}\mathrm{platform}& \mathrm{Newly}\mathrm{built}\mathrm{platform}\end{array}\right)\\ M_{133}=\left(\begin{array}{lll}{M}_{133},& \mathrm{Platform}\mathrm{Function},& \mathrm{Collection}\oplus \mathrm{Statistics}\oplus \mathrm{Upload}\end{array}\right)\end{array}\right.\end{array}\right.$$

(18)

$${L_1}^{\prime }=\left\{\begin{array}{c}{T}_{11}{L}_1=L_{11}=\left[\begin{array}{ccc}{M}_1^{\prime },& \mathrm{Information}\mathrm{traceability},& 4\\ & \mathrm{Response}\mathrm{speed},& 1\mathrm{h}\end{array}\right]\hfill \\ T_{12}{L}_1=L_{12}=\left[\begin{array}{ccc}{M}_1^{″},& \mathrm{Information}\mathrm{traceability},& 5\\ & \mathrm{Response}\mathrm{speed},& 0.1\mathrm{h}\end{array}\right]\hfill \end{array}\right.$$

(19)

With the help of e-commerce platform $L_{11}$ for information exchange, the correlation degree of information traceability $K_{111}=0.5$ and the correlation degree of response time $K_{121}=0.67$. Due to the use of an existing platform, there is no new carbon emission $C_{11}=0$. The correlation degree $K_{112}$ of the rebuilt platform $L_{12}$ is 1, $K_{122}\approx 0.2$, and the carbon emission $C_{12}>0$. The correlation degree of $L_{11}$ and $L_{12}$ and the target $G_1$ are positive, that is, $G_1\downarrow {L_1}^{\prime }$, which can be used as an alternative solution basic-element. Bringing the two divergence matter-elements into the actual situation can form two schemes: ① directly use the existing e-commerce platform to collect, count and upload residents’ needs; ② The government will build a new intelligent demand interaction platform.

4.2. MI₂ Professionalization

Cold chain logistics is highly professional. The volunteers and doctors responsible for the transportation of fresh products are limited by their own experience and lack the professional knowledge of cold chain logistics transportation. Due to the lack of professional guidance and non-standard transportation operation, it is very easy to cause food pollution and deterioration, as well as virus contaminated food materials, leading to the spread of the pandemic. In addition, due to the need of transportation and movement, the transportation personnel may be infected if they contact the patients in the incubation period during their work. At present, the probability of accident risks such as food pollution, personnel infection or transportation failure caused by non-standard transportation operation of volunteers is 5%. In order to ensure the safety and cost of pandemic prevention activities, the probability of transportation risk should be less than 1%, and the best advantage is 0.5%. The conditions matter-element and target affair-element are expressed as:

$$L_2=\left[M_2,\mathrm{Transportation}\mathrm{Risk},5\%\right]$$

(20)

$$G_2=\left[\begin{array}{ccc}\mathrm{distribute},& \mathrm{Dominated}\mathrm{target},& \mathrm{Fresh}\mathrm{Products}\\ & \mathrm{Risk},& <0,1\%>\end{array}\right]$$

(21)

The matter-element of the $distributors$ M2 can be expressed as:

$$M_2=\left(M_2,\mathrm{Distributors},\mathrm{Volunteers}\right)$$

(22)

According to model (6), the correlation degree of the new target and condition is $K_2=-8$, and the problem model is recorded as $G_2\uparrow L_2$. In order to meet the needs of cold chain transportation and improve the safety of fresh products in pandemic prevention areas, the $M_2$ matter element is divergent.

$$M_2⊣\left\{\begin{array}{c}{M}_{21}=\left(\begin{array}{ccc}{M}_{21},& \mathrm{Distributors},& \mathrm{Trained}\mathrm{volunteers}\end{array}\right)\hfill \\ M_{22}=\left(\begin{array}{ccc}{M}_{22},& \mathrm{Distributors},& \mathrm{Automated}\mathrm{Supply}\mathrm{Chain}\end{array}\right)\hfill \\ M_{23}=\left(\begin{array}{ccc}{M}_{23},& \mathrm{Distributors},& \mathrm{Volunteer}\oplus \mathrm{Professional}\mathrm{Logistician}\end{array}\right)\hfill \end{array}\right.$$

(23)

Similarly, in combination with the safety and professional needs of residents’ fresh food transportation during the pandemic prevention period, the following divergent elements are selected as measure elements:

$${L_2}^{\prime }=\left\{\begin{array}{l}{T}_{21}{L}_2=L_{21}=\left[\begin{array}{lll}{M}_{21},& \mathrm{Transportation}\mathrm{Risk},& 0.3\%\end{array}\right]\\ T_{22}{L}_2=L_{22}=\left[\begin{array}{lll}{M}_{22},& \mathrm{Transportation}\mathrm{Risk},& 0.1\%\end{array}\right]\\ T_{23}{L}_2=L_{23}=\left[\begin{array}{lll}{M}_{23},& \mathrm{Transportation}\mathrm{Risk},& 0.6\%\end{array}\right]\end{array}\right.$$

(24)

If the trained volunteer $M_{21}$ is arranged for transportation, and the correlation degree $K_{21}=0.6$, the training activity will produce carbon emission $C_{21}$. If an automated supply chain $M_{22}$ is established for transshipment activities, it is highly professional and has a low transport risk. The correlation degree $K_{22}=0.2$, but the construction of the supply chain will produce fixed carbon emission $c_{221}$. The carbon emission generated by mechanical operation is equal to the closure date $t_1$ multiplied by the single day carbon emission $c_{222}$, and the total carbon emission is $C_{22}$. Compared with $M_{22}$, the transport risk of $M_{23}$ is slightly higher, and the correlation degree $K_{23}=0.8$, but the carbon emission of the guidance activity is very low, so it is ignored. According to the divergent matter-element results, the solutions are as follows: ③ the government trains volunteers to reduce the transportation risk caused by non-standard operation; ④ it establishes an automatic transportation supply chain; ⑤ professional logisticians guide volunteers to carry out transportation operations.

4.3. MI₃ Optimizing Transit Efficiency

The contradiction between the rapidly rising transportation demand during the pandemic prevention period and the lack of cold chain logistics infrastructure is difficult to be changed in a short time, and the construction of cold chain logistics facilities that meet the needs cannot be completed in a short time. Under the current transportation conditions, the transport capacity of cold chain transport vehicles is limited. If ordinary freight vehicles are used to transport fresh food, the food may be polluted in the process of transportation, loading and unloading, and if the climate is hot, it will lead to food deterioration. However, only refrigerated vehicles with limited transport capacity cannot meet the large number of food distribution needs of urban residents. This can be expressed as follows:

$$P=\left(G_{31}\wedge G_{32}\right)\ast L_3$$

(25)

$$\left(G_{31}\wedge G_{32}\right)\ast L_3\uparrow L_3$$

(26)

$G_{31}$ and $G_{32}$ are the two transportation targets, respectively. $G_{31}$ is to transport enough fresh food, and $G_{32}$ is to ensure the transportation temperature of fresh food to avoid decay. L indicates the existing cold chain infrastructure conditions. There are 24 residential buildings in the case community, with a population of about 10,000 people, and the daily demand for fresh food per capita is 0.5–1 kg. Therefore, taking 3 days as a distribution cycle, the total daily demand for fresh food in the community is 15–30 t, the optimal value is 30 t, the transportation temperature of fresh products is 0–7 °C, the optimal value is 2 °C, and the transportation distance of cold storage is 15 km. Although the transportation capacity of general trucks is sufficient, it cannot meet the needs of transportation temperature, and an insufficient number of refrigerator cars leads to the shortage of transport capacity, which can only be allocated to the community for three round trips in a single day, i.e., 90 km and 9 t in total. Expressed by basic-element as:

$$G_{31}=\left[\begin{array}{ccc}\mathrm{Transport},& \mathrm{Receiving}\mathrm{object},& \mathrm{Fresh}\mathrm{Products}\\ & \mathrm{Total}\mathrm{weight},& <15\mathrm{t},30\mathrm{t}>\end{array}\right]$$

(27)

$$G_{32}=\left[\begin{array}{ccc}\mathrm{Transport},& \mathrm{Receiving}\mathrm{object},& \mathrm{Fresh}\mathrm{Products}\\ & \mathrm{temperature},& <0,7°\mathrm{C}>\end{array}\right]$$

(28)

$$L_3=\left[\begin{array}{c}\begin{array}{ccc}\mathrm{Infrastructure}{M}_3,& \mathrm{Total}\mathrm{weight},& 9\mathrm{t}\\ & \mathrm{Temperature},& 2°\mathrm{C}\end{array}\end{array}\right]$$

(29)

Use correlation analysis to $M_3$ matter-element:

$$M_3=\left\{\begin{array}{l}{M}_{31}=\left(\begin{array}{ccc}\mathrm{Refrigerator}\mathrm{car}{M}_{31},& \mathrm{Carrying}\mathrm{capacity},& 3\mathrm{t}\\ & \mathrm{Carrying}\mathrm{temperature},& 2°\mathrm{C}\end{array}\right)\\ M_{32}=\left(\begin{array}{ccc}\mathrm{Cold}\mathrm{storage}{M}_{32},& \mathrm{Distance},& 15\mathrm{km}\end{array}\right)\end{array}\right.$$

(30)

According to model (8) and model (5), when using a refrigerator car, the dependent degree of targets $G_{31}$ and $G_{32}$ and conditions $k_{31}=-0.2$, $k_{32}=1$, that is, $\left(G_{31}\wedge G_{32}\right)\uparrow L_3$. Due to the shortage of refrigerator cars and large transportation demand, some fresh food can only be transported by ordinary trucks. This kind of “chain breakage” phenomenon of cold chain logistics leads to low food quality and safety. In order to meet the needs of cold chain transportation and improve the safety of fresh products in pandemic prevention areas, we use divergence analysis on the matter-element of $M_{31}$ $M_{32}$:

$$M_{31}⊣\left\{\begin{array}{l}{M}_{311}=\left(\begin{array}{lll}\mathrm{Car}{M}_{311},& \mathrm{Carrying}\mathrm{capacity},& 0.5\mathrm{t}\\ & \mathrm{Carrying}\mathrm{temperature},& 26°\mathrm{C}\end{array}\right)\\ M_{312}=\left(\begin{array}{lll}\mathrm{Truck}{M}_{312},& \mathrm{Carrying}\mathrm{capacity},& 10\mathrm{t}\\ & \mathrm{Carrying}\mathrm{temperature},& 26°\mathrm{C}\end{array}\right)\\ M_{313}=\left(\begin{array}{lll}\mathrm{Modified}\mathrm{truck}{M}_{313},& \mathrm{Carrying}\mathrm{capacity},& 10\mathrm{t}\\ & \mathrm{Carrying}\mathrm{temperature},& 6°\mathrm{C}\end{array}\right)\hfill \\ \dots \end{array}\right.$$

(31)

$$M_{32}⊣\left\{\begin{array}{c}{M}_{321}=\left(\begin{array}{ccc}\mathrm{Newly}\mathrm{built}\mathrm{cold}\mathrm{storage}{M}_{322},& \mathrm{Distance},& 5\mathrm{km}\end{array}\right)\hfill \\ M_{322}=\left(\begin{array}{ccc}\mathrm{Nearby}\mathrm{warehouses}{M}_{322},& \mathrm{Distance},& 5\mathrm{km}\end{array}\right)\hfill \\ M_{323}=\left(\begin{array}{ccc}\mathrm{Supermarket}\mathrm{warehouse}{M}_{323},& \mathrm{Distance},& 7.5\mathrm{km}\end{array}\right)\hfill \\ \dots \hfill \end{array}\right.$$

(32)

The following divergent elements are selected as measure elements:

$$T{M}_3=\left\{\begin{array}{l}{M_3}^{\prime }=\left\{\begin{array}{l}{M}_{31}=\left(\begin{array}{lll}\mathrm{Refrigerator}\mathrm{car}{M}_{31},& \mathrm{Carrying}\mathrm{capacity},& 3\mathrm{t}\\ & \mathrm{Carrying}\mathrm{temperature},& 2°\mathrm{C}\end{array}\right)\\ M_{322}=\left(\begin{array}{lll}\mathrm{Nearby}\mathrm{warehouses}{M}_{322},& \mathrm{Distance},& 5\mathrm{km}\end{array}\right)\end{array}\right.\\ {M_3}^{″}=\left\{\begin{array}{l}{M}_{313}=\left(\begin{array}{lll}\mathrm{Modified}\mathrm{truck}{M}_{313},& \mathrm{Carrying}\mathrm{capacity},& 10\mathrm{t}\\ & \mathrm{Carrying}\mathrm{temperature},& 6°\mathrm{C}\end{array}\right)\\ M_{322}=\left(\begin{array}{lll}\mathrm{Cold}\mathrm{storage}{M}_{32},& \mathrm{Distance},& 15\mathrm{km}\end{array}\right)\end{array}\right.\end{array}\right.$$

(33)

In the actual environment, we can get measures ⑥: Use the nearby warehouses to establish a closer temporary cold storage $M_{322}$ as the fresh food transit station, shorten the distance to one third, and refrigerated vehicles can transport nine times, with a total transport capacity of 27 t in a distribution cycle. Measure ⑦: Modify ordinary trucks to obtain the modified truck $M_{313}$ that meets the conditions of cold chain transportation. Since there are enough ordinary trucks, the total transport capacity can be expressed by the maximum demand of 30 t:

$${L_3}^{\prime }=\left\{\begin{array}{l}{T}_{31}{L}_3=\left\{\begin{array}{l}\left(\begin{array}{lll}\mathrm{Infrastructure}{M}_3^{\prime },& \mathrm{Carrying}\mathrm{capacity},& 27\mathrm{t}\\ & \mathrm{Carrying}\mathrm{temperature},& 2°\mathrm{C}\end{array}\right)\end{array}\right.\\ T_{32}{L}_3=\left\{\begin{array}{l}\left(\begin{array}{lll}\mathrm{Infrastructure}{M}_3^{″},& \mathrm{Carrying}\mathrm{capacity},& 30\mathrm{t}\\ & \mathrm{Carrying}\mathrm{temperature},& 6°\mathrm{C}\end{array}\right)\end{array}\right.\end{array}\right.$$

(34)

When establishing the temporary cold storage using nearby warehouses $M_{322}$, the correlation degree for the objectives $G_{31}$ and $G_{32}$ is $K_{311}=0.8$ $K_{321}=1$. In this measure, the carbon emission $C_{31}$ is equal to the carbon emission of the cold chain transport vehicle $c_{311}$ plus the carbon emission produced by the refrigeration of the temporary cold storage $c_{312}$. $c_{311}$ is the closure date $t_1$ multiplied by the transportation distance $m$ multiplied by the carbon emission coefficient per 100 km of the cold chain transport vehicle ${\alpha }_1$ divided by the transportation cycle $t_2$, $c_{312}$ is the closed date t multiplied by the carbon emission $c_{3121}$ of the daily refrigeration electricity of the cold storage (the carbon emission generated by vehicle refrigeration during loading and unloading is ignored). When transporting fresh food by refitting the general truck $M_{313}$, the correlation degree of the target $G_{31}$ and $G_{32}$ is $K_{312}=0.8$ $K_{322}=0.8$, and the dependent function value is positive. The carbon emission $C_{32}$ of this measure is equal to the closure date $t_1$ multiplied by the transportation distance $m$ multiplied by the refrigeration carbon emission per 100 km of the refitted truck ${\alpha }_2$. As the refrigeration vehicle is a modified truck, the carbon emission must be greater than that of the professional cold chain transport vehicle ${\alpha }_2$ > ${\alpha }_1$. According to the results of matter-element divergence, we can get measure ⑥ to establish temporary cold storage using nearby warehouses or other facilities to shorten the distribution distance and measure ⑦ to modify trucks to increase the carrying capacity.

4.4. MI₄ Contactless Delivery

After the fresh products are transported to residential areas, how to deliver them to residents has become the last problem to ensure fresh products are delivered to residents in pandemic prevention areas. In this case, residents go out to pick up goods by themselves in a fixed period of time. This method does not separate the pick-up time and needs contact with the administrator, so it is difficult to avoid the risk of cross infection. Assuming that each resident will encounter at least one administrator and two other pickup residents in this pickup process, the average person contact of residents under the pandemic prevention measures is 6, and the positive field of person contact of the target is < 0,2 >, the discrete correlation function, target and condition basic-element of the contradiction problem are expressed as follows:

$$K_{42}=k\left(x\right)=\left\{\begin{array}{c}1,x=0\\ 0.5,x=1\\ 0,x=2\\ -0.5,x=3\\ \dots \end{array}\right.$$

(35)

$$G_4=\left[\begin{array}{ccc}\mathrm{Deliver},& \mathrm{Dominated}\mathrm{target},& \mathrm{Fresh}\mathrm{Products}\\ & \mathrm{Personnel}\mathrm{contact},& 0\end{array}\right]$$

(36)

$$L_4=\left[\begin{array}{ccc}{M}_4,& \mathrm{Function},& \mathrm{Delivery}\\ & \mathrm{Personnel}\mathrm{contact},& 5\end{array}\right]$$

(37)

Divergence analysis of the compound affair-element $M_4$ of the distribution function:

$$M_4=\left\{\begin{array}{l}{M}_{41}=\left(\begin{array}{lll}\mathrm{Deliver},& \mathrm{Actuating}\mathrm{object},& \mathrm{Volunteer}\end{array}\right)\\ M_{42}=\left(\begin{array}{lll}\mathrm{Pick}\mathrm{up},& \mathrm{Place},& \mathrm{Community}\mathrm{gate}\end{array}\right)\\ M_{43}=\left(\begin{array}{lll}\mathrm{Facilities},& \mathrm{Disinfection}\mathrm{function},& \mathrm{None}\end{array}\right)\end{array}\right.$$

(38)

According to model (35), the correlation degree of target and condition $K_4=-1$. It is an incompatibility. In order to reduce the personal contact of ordinary residents in the process of pandemic prevention, the divergence analysis of basic-element $M_{41}$ $M_{42}$ $M_{43}$:

$$M_{41}⊣\left\{\begin{array}{l}{M}_{411}=\left(\begin{array}{lll}\mathrm{Deliver},& \mathrm{Actuating}\mathrm{object},& \mathrm{Administrator}\end{array}\right)\\ M_{412}=\left(\begin{array}{lll}\mathrm{Deliver},& \mathrm{Actuating}\mathrm{object},& \mathrm{UAV}\end{array}\right)\\ M_{413}=\left(\begin{array}{lll}\mathrm{Deliver},& \mathrm{Actuating}\mathrm{object},& \mathrm{Delivery}\mathrm{robots}\end{array}\right)\end{array}\right.$$

(39)

$$M_{42}⊣\left\{\begin{array}{l}{M}_{421}=\left(\begin{array}{lll}\mathrm{Pick}\mathrm{up},& \mathrm{Place},& \mathrm{Apartment}\mathrm{entrance}\end{array}\right)\\ M_{422}=\left(\begin{array}{lll}\mathrm{Pick}\mathrm{up},& \mathrm{Place},& \mathrm{Elevator}\mathrm{entrance}\end{array}\right)\\ M_{423}=\left(\begin{array}{lll}\mathrm{Pick}\mathrm{up},& \mathrm{Place},& \mathrm{Floor}\mathrm{access}\end{array}\right)\\ M_{424}=\left(\begin{array}{lll}\mathrm{Pick}\mathrm{up},& \mathrm{Place},& \mathrm{Resident}\mathrm{door}\end{array}\right)\end{array}\right.$$

(40)

$$M_{43}⊣\left\{\begin{array}{l}{M}_{431}=\left(\begin{array}{lll}\mathrm{Facilities},& \mathrm{Disinfection}\mathrm{function},& \mathrm{High}\text{–}\mathrm{temperature}\mathrm{disinfection}\end{array}\right)\\ M_{432}=\left(\begin{array}{lll}\mathrm{Facilities},& \mathrm{Disinfection}\mathrm{function},& \mathrm{Chemical}\mathrm{disinfection}\end{array}\right)\\ M_{433}=\left(\begin{array}{lll}\mathrm{Facilities},& \mathrm{Disinfection}\mathrm{function},& \mathrm{Ultraviolet}\mathrm{disinfection}\end{array}\right)\\ M_{434}=\left(\begin{array}{lll}\mathrm{Facilities},& \mathrm{Disinfection}\mathrm{function},& \mathrm{Vacuum}\mathrm{packing}\end{array}\right)\end{array}\right.$$

(41)

$$T{M}_4=\left\{\begin{array}{l}{M_4}^{\prime }=\left\{\begin{array}{l}{M}_{41}=\left(\begin{array}{ccc}\mathrm{Deliver},& \mathrm{Actuating}\mathrm{object},& \mathrm{Volunteer}\end{array}\right)\\ M_{424}=\left(\begin{array}{ccc}\mathrm{Pick}\mathrm{up},& \mathrm{Place},& \mathrm{Resident}\mathrm{door}\end{array}\right)\\ M_{433}=\left(\begin{array}{ccc}\mathrm{Facilities},& \mathrm{Disinfection}\mathrm{Function},& \mathrm{Vacuum}\mathrm{packing}\end{array}\right)\end{array}\right.\\ {M_4}^{″}=\left\{\begin{array}{l}{M}_{413}=\left(\begin{array}{lll}\mathrm{Deliver},& \mathrm{Actuating}\mathrm{object},& \mathrm{Delivery}\mathrm{robots}\end{array}\right)\\ M_{424}=\left(\begin{array}{lll}\mathrm{Pick}\mathrm{up},& \mathrm{Place},& \mathrm{Resident}\mathrm{door}\end{array}\right)\\ M_{433}=\left(\begin{array}{lll}\mathrm{Facilities},& \mathrm{Disinfection}\mathrm{function},& \mathrm{Ultraviolet}\mathrm{disinfection}\end{array}\right)\end{array}\right.\\ {M_4}^{‴}=\left\{\begin{array}{l}{M}_{411}=\left(\begin{array}{ccc}\mathrm{Deliver},& \mathrm{Actuating}\mathrm{object},& \mathrm{Administrator}\end{array}\right)\\ M_{322}=\left(\begin{array}{ccc}\mathrm{Pick}\mathrm{up},& \mathrm{Place},& \mathrm{Floor}\mathrm{access}\end{array}\right)\\ M_{322}=\left(\begin{array}{ccc}\mathrm{Facilities},& \mathrm{Disinfection}\mathrm{Function},& \mathrm{Ultraviolet}\mathrm{disinfection}\end{array}\right)\end{array}\right.\end{array}\right.$$

(42)

$$T{L}_4=\left\{\begin{array}{c}{L_4}^{\prime }=\left[\begin{array}{ccc}{M}_4^{\prime },& \mathrm{Function},& \mathrm{Delivery}\\ & \mathrm{Personnel}\mathrm{contact},& 0\end{array}\right]\\ {L_4}^{″}=\left[\begin{array}{ccc}{M}_4^{″},& \mathrm{Function},& \mathrm{Delivery}\\ & \mathrm{Personnel}\mathrm{contact},& 0\end{array}\right]\\ {L_4}^{‴}=\left[\begin{array}{ccc}{M}_4^{‴},& \mathrm{Function},& \mathrm{Delivery}\\ & \mathrm{Personnel}\mathrm{contact},& 1\end{array}\right]\end{array}\right.$$

(43)

The obtained composite basic-elements are substituted into the actual environment to obtain measures: ⑧ Volunteers will deliver fresh products to the door of each household and use plastic vacuum packaging to avoid food pollution. Residents will open the door to pick up the goods after the volunteers leave. The number of personnel contacts of the measure is zero. According to model (35), the correlation degree of target and condition is $k_{41}=1$. The carbon emission $C_{41}$ generated by this measure is equal to the carbon emission $c_{41}$ of manufacturing plastic vacuum packaging every day multiplied by the number of closed days $t_1$. Measure ⑨: Delivery robots with an ultraviolet disinfection lamp are used to distribute fresh products to the residents’ door. At this time, the number of personnel contacts is zero, and the correlation degree of target and condition is $k_{42}=1$. The carbon emission $C_{42}$ generated by this measure is the carbon emission $c_{42}$ of the power consumed by the distribution robot for daily charging multiplied by the closed days $t_1$. Measure ⑩:The administrator delivers fresh products to the storage cabinets on each floor, and installs ultraviolet lights in the storage cabinets for disinfection. Residents go to the storage cabinets on their own floors to pick up goods. This measure may only produce personnel contact when residents on the same floor pick up goods, and the average number of personnel contacts is reduced to one. The correlation degree of this measure is $k_{43}=0.5$. The carbon emission $C_{43}$ is equal to the carbon emission $c_{43}$ generated by the daily power consumption of the disinfection lamp multiplied by the closed days $t_1$.

4.5. Superiority Evaluation

After the alternative measures are combined into the alternative scheme, the better scheme can be found through the method of excellence evaluation. According to the priority of different stages, the weight coefficients of stages MI₁ MI₂ MI₂ MI₄ are respectively ${\beta }_1=0.3$ ${\beta }_2=0.2$ ${\beta }_3=0.2$ ${\beta }_4=0.3$. After normalizing the correlation degree of each scheme, the corresponding superiority degree of each scheme can be calculated according to model (10), the normative correlation degree and carbon emission of measures are reported in

Table 1. Normative correlation degree and carbon emission.

Stage	Weight	Measure	Correlation Degree K	$\mathbf{Normative}\mathbf{Correlation}\mathbf{Degree}{\mathit{K}}^{\prime }$	Superiority Degree	$\mathbf{Carbon}\mathbf{Emission}\mathit{C}$
MI1	0.3	①	0.5	0.5	0.45	0
			0.67	1
		②	1	1	0.39	$C_{12}$
			0.2	0.3
MI2	0.2	③	0.6	0.75	0.15	$C_{21}$
		④	0.2	0.25	0.05	$C_{22}=c_{221}+t_1{c}_{222}$
		⑤	0.8	1	0.2	0
MI3	0.2	⑥	0.8	0.8	0.36	$C_{31}=t_{1}m{\alpha }_1/t_2+t_1{c}_{3121}$
			1	1
		⑦	1	1	0.24	$C_{32}=t_{1}m{\alpha }_2$
			0.2	0.2
MI4	0.3	⑧	1	1	0.3	$C_{41}=t_{1}g{c}_{41}$
		⑨	1	1	0.3	$C_{42}=t_1{c}_{42}$
		⑩	0.5	0.5	0.15	$C_{43}=t_1{c}_{43}$

.

There are 10 measures in four stages, and 36 schemes, $Z_i$ can be obtained. The superiority degree of the scheme is obtained according to model (10), as shown in

Table 2. Scheme superiority degree.

Scheme	Measures	$\mathbf{Superiority}\mathit{G}$	Scheme	Measures	$\mathbf{Superiority}\mathit{G}$
$Z_1$	①③⑥⑧	1.26	$Z_{19}$	②③⑥⑧	1.2
$Z_2$	①③⑥⑨	1.26	$Z_{20}$	②③⑥⑨	1.2
$Z_3$	①③⑥⑩	1.11	$Z_{21}$	②③⑥⑩	1.05
$Z_4$	①③⑦⑧	1.14	$Z_{22}$	②③⑦⑧	1.08
$Z_5$	①③⑦⑨	1.14	$Z_{23}$	②③⑦⑨	1.08
$Z_6$	①③⑦⑩	0.99	$Z_{24}$	②③⑦⑩	0.93
$Z_7$	①④⑥⑧	1.16	$Z_{25}$	②④⑥⑧	1.1
$Z_8$	①④⑥⑨	1.16	$Z_{26}$	②④⑥⑨	1.1
$Z_9$	①④⑥⑩	1.01	$Z_{27}$	②④⑥⑩	0.95
$Z_{10}$	①④⑦⑧	1.04	$Z_{28}$	②④⑦⑧	0.98
$Z_{11}$	①④⑦⑨	1.04	$Z_{29}$	②④⑦⑨	0.98
$Z_{12}$	①④⑦⑩	0.89	$Z_{30}$	②④⑦⑩	1.25
$Z_{13}$	①⑤⑥⑧	1.31	$Z_{31}$	②⑤⑥⑧	1.25
$Z_{14}$	①⑤⑥⑨	1.31	$Z_{32}$	②⑤⑥⑨	0.83
$Z_{15}$	①⑤⑥⑩	1.16	$Z_{33}$	②⑤⑥⑩	1.1
$Z_{16}$	①⑤⑦⑧	1.19	$Z_{34}$	②⑤⑦⑧	1.13
$Z_{17}$	①⑤⑦⑨	1.19	$Z_{35}$	②⑤⑦⑨	1.13
$Z_{18}$	①⑤⑦⑩	1.04	$Z_{36}$	②⑤⑦⑩	0.98

.

The four schemes $Z_{13}$, $Z_{14}$, $Z_1$ and $Z_2$ with the highest superiority are selected as the optional schemes of emergency cold chain logistics. In order to reduce carbon emissions and achieve higher energy efficiency, the total carbon emissions of these four schemes are calculated to select the scheme with better carbon emissions.

According to the measures of different schemes, the total carbon emission model of each scheme can be obtained:

$$C\left(Z_{13}\right)=t_{1}m{\alpha }_1/t_2+t_1{c}_{3121}+t_1{c}_{41}$$

(44)

$$C\left(Z_{14}\right)=t_{1}m{\alpha }_1/t_2+t_1{c}_{3121}+t_1{c}_{42}$$

(45)

$$C\left(Z_1\right)=c_{21}+t_{1}m{\alpha }_1/t_2+t_1{c}_{3121}+t_1{c}_{41}$$

(46)

$$C\left(Z_2\right)=c_{21}+t_{1}m{\alpha }_1/t_2+t_1{c}_{3121}+t_1{c}_{42}$$

(47)

Assuming that the community is closed for $t_1=21$ days, the transportation cycle is $t_1=3$ days, and the fixed carbon emission of training activities is $C_{21}=10\mathrm{kg}$, the total transportation distance is $m=90\mathrm{km}$, the fuel consumption of a small refrigerator car is about 12 L/100 km, the carbon emission per liter of diesel is about 2.63 kg, and the carbon emission coefficient of a refrigerator car under the standard load is ${\alpha }_1=31.56$. A 5 m² cold storage that can hold 30 t goods consumes 34.5 kWh of electricity per day, while the temporary cold storage needs to consume more electricity to ensure the refrigeration temperature due to its poor insulation capacity and refrigeration efficiency. Therefore, assuming that the daily electricity consumption is 1.5 times that of the standard cold storage, that is 51.75 kwh. The carbon emission per kWh is 0.785 kg, and the carbon emission of daily refrigeration electricity of the cold storage $c_{3121}=40.62\mathrm{kg}$. In measure ⑥, the daily supply of fresh products is $g=9\mathrm{t}$. One plastic bag can hold 1.5 kg of fresh products and produce about 0.1 g carbon emission. In this case, the carbon emission produced by using plastic vacuum packaging every day is $c_{41}=0.6\mathrm{kg}$; There are 24 residential apartments in total. Each delivery robot can complete the transportation of fresh food in one residential apartment in one day. In one distribution cycle, it needs to complete the distribution of at least eight residential apartments every day, so it needs eight delivery robots. Taking the power of a delivery robot of a brand as an example, its battery capacity is 20 Ah and the voltage is 29.4 V. If the energy loss in the charging process is not considered, it needs 0.588 kWh to be fully charged. The automatic distribution robot is charged once a day, and the daily charging of eight machines produces carbon emission $c_{42}\approx 3.69\mathrm{kg}$.

According to model (43)–(46):$C\left(Z_{13}\right)=1064.448\mathrm{kg},$ $C\left(Z_{14}\right)=1129.338\mathrm{kg}$, $C\left(Z_1\right)=1074.448\mathrm{kg}$, $C\left(Z_2\right)=1139.338$ kg. Scheme $Z_{13}$ has the lowest carbon emissions, so the following measures can be selected: use the e-commerce platform to collect, statistic and upload residents’ needs to improve the efficiency of information transmission; arrange professional logisticians to guide volunteers to carry out transportation, loading and unloading operations to reduce transportation risks; use the nearby warehouses to establish a temporary cold storage to shorten the distribution distance; and use plastic vacuum packaging and contactless distribution to avoid resident contact.

Integrating or directly using the e-commerce platform of local large supermarkets eliminates the cost and labor of establishing the platform and can effectively convey the residents’ demand for fresh products. The transmission of demand does not need to go through the intermediate link of volunteers, but directly reaches the supplier, which greatly increases the efficiency of information transmission and reduces the workload of volunteers. The supplier’s e-commerce platform can more comprehensively display the types of commodity supply, and the loss caused by the expiration of some commodities due to the lack of promotion channels. On the other hand, with the help of the e-commerce platform, it can accurately check and trace the transportation path and batch of products, so as to facilitate the tracing of product sources and other information. The guidance of professional logisticians can provide transportation and storage facilities and plan distribution routes through logistics enterprises and can make up for the loss and pollution of fresh products caused by a lack of professional knowledge. Compared with the establishment of an automatic supply chain, this measure has low cost and a short investment cycle, and can avoid the carbon emission caused by the extensive use of machinery, which is more suitable for the urgent, low-carbon and efficient needs of emergency disaster relief. Using large warehouses to establish temporary cold storage is more convenient and quicker than transforming trucks, which can be put into use in time in the early stage of the disaster, ensuring the stable supply of residents’ fresh products and avoiding the phenomenon of “chain breaking”. Fresh products are packaged in vacuum plastic and sent to the door of residents by volunteers without contact, which also avoids residents’ contact on the basis of ensuring feasibility and low carbon as far as possible. The scheme solves the contradiction of cold chain logistics of fresh products in the case, optimizes all stages, and controls the total carbon emission on the premise of high optimization of the scheme.

5. Conclusions

This paper studies the method of generating strategies for solving the dilemma of cold chain logistics in emergencies, as well as the evaluation and selection of strategies. We use the formal logic of extension basic-element to decompose and analyze the contradictions in the four links of emergency cold chain logistics. With the help of the extensible analysis method of extenics, 10 strategic elements with a positive dependent function are generated, which are combined to form 36 feasible strategies, and the superiority evaluation method is used to rank and evaluate the strategies. In the process of scheme selection, we fully consider the applicability of emergency rescue and add the idea of carbon emission control, which not only ensures the effective operation of emergency logistics, but also avoids excessive carbon emission and energy waste. Finally, the specific plan of integrating an e-commerce platform, expert guidance, establishing temporary cold storage transfer and contactless distribution is designed. This research helps to serve as a reference in the global long term COVID-19 situation, as well as other emergency relief environments, and also provides direction for low-carbon and efficient development of the supply chain industry. According to the disaster situation and the limitations of local conditions, the emergency logistics in the disaster area must design the scheme specifically in order to ensure the timeliness and safety of logistics transportation. If we can use the formal language of extenics and design emergency logistics scheme with the help of computer and artificial intelligence, it will not only effectively support the disaster relief activities, but also greatly improve the confidence of the people in the disaster area. It is of great significance to the development of emergency logistics in the future. At present, the combination of extenics and artificial intelligence is still in its infancy. With the further development of research, extenics, as a cross-sectional discipline, will become an efficient means and thinking paradigm for programmatically solving contradictions in various fields.

In addition, this paper has two limitations. On the one hand, due to the complexity of the emergency rescue environment, the case analyzed in this paper is the emergency food supply chain rather than the complete emergency cold chain logistics, which cannot fully show the application of extenics in the generation of emergency cold chain logistics strategies. On the other hand, although this paper refers to PRISMA statement as the method of literature review, it is not fully carried out according to the paradigm of the PRISMA statement due to the limitation of the length of the article, which needs to be further expanded in future research.