Ecological Eciency Evaluation and Spatiotemporal Characteristics Analysis of the Coupling Coordination of the Logistics Industry and Manufacturing Industry

10 The coupling coordination of the logistics industry and manufacturing industry conducive to the 11 sustainable development of logistics and manufacturing and the stability of sustainable supply chain. 12 The logistics and manufacturing industries are not only the basic industries that support social 13 development, but also the industries with high carbon emissions. Firstly, this paper classifies the 14 carbon emissions from the logistics industry and manufacturing industry as undesirable outputs, 15 evaluates the ecological efficiency of the logistics industry (LEE) and manufacturing industry (MEE) 16 in the Yangtze River Delta from 2006 to 2019 by using the unexpected slacks-based measure (SBM) 17 model. Secondly, the coupling coordination method is used to analyze the coupling coordination 18 scheduling of industrial ecological efficiency. Thirdly, the paper analyzes the spatial differences of 19 the coupling coordination ecological efficiency between logistics industry and manufacturing 20 industry (MLCC) by using the exploratory spatial data analysis method. Finally, the spatial 21 econometric model is used to analyze the driving factors of the MLCC. The results show: The 22 ecological efficiency of the manufacturing industry has steadily improved. The ecological efficiency 23 of the logistics industry presents the rising trend in fluctuation. The level of the coupling coordination 24 development between the logistics and manufacturing industries is high. The results of the spatial 25 heterogeneity analysis show that the spatial differentiation of high-high agglomeration and low-low 26 agglomeration is obvious. The spatial agglomeration characteristics are relatively stable, and the 27 spatial diffusion effect is strong; In space, the MLCC shows a trend of developing from multiple 28 agglomeration areas to one agglomeration area. The results of driving factor analysis show that 29 foreign direct investment(FDI), government intervention(GI) and human capital(HP) have positive 30 effects on the MLCC, while industrial structure(IS), environmental regulation(ER) and energy 31 intensity(EI) have negative effects on the MLCC. 32 model to measure the of the panel Tobit model used to empirically analyze its influencing factors. The results show that the overall efficiency of China's logistics industry is at the upper middle level, with obvious regional differences. nonmandatory stochastic technology panel The show that the overall efficiency of green technology innovation the national still low, the capital intensity and industry profit rate have a significantly positive effect on the efficiency of green technology innovation in the advanced industry. Using the directional distance function, the ecological efficiency level of the manufacturing is calculated under the condition of considering the energy input and undesirable outputs. The research shows that the ecological efficiency of China's is on the rise, but it is significantly lower than that without considering and negative The ecological efficiency level of the ecological efficiency of the coupling coordination development of industry and manufacturing industry considering environmental effects, as well the spatial model (spatial heterogeneity, spatial agglomeration and driver analysis) and trend dynamic evolution of the coupling coordination development efficiency of the two industries. The contributions of this paper are as follows: (1) The coupling coordination measurement system of logistics industry and manufacturing industry considering carbon emission is established. (2) This paper analyzes the spatial effect of the coupling coordination degree between logistics industry and manufacturing industry by using spatial correlation model, and discusses the influencing factors of coupling coordination by using spatial econometric model. The analysis has certain policy guidance and theoretical significance. The rest of this study is as follows. The second part presents introduces the relevant models. The third part explains the indicators and unexpected data. The fourth part presents the empirical analysis. Finally, this paper summarizes the conclusions and discusses relevant suggestions.

The coupling coordination of the logistics industry and manufacturing industry conducive to the 11 sustainable development of logistics and manufacturing and the stability of sustainable supply chain. 12 The logistics and manufacturing industries are not only the basic industries that support social 13 development, but also the industries with high carbon emissions. Firstly, this paper classifies the 14 carbon emissions from the logistics industry and manufacturing industry as undesirable outputs, 15 evaluates the ecological efficiency of the logistics industry (LEE) and manufacturing industry (MEE) 16 in the Yangtze River Delta from 2006 to 2019 by using the unexpected slacks-based measure (SBM) 17 model. Secondly, the coupling coordination method is used to analyze the coupling coordination 18 scheduling of industrial ecological efficiency. Thirdly, the paper analyzes the spatial differences of 19 the coupling coordination ecological efficiency between logistics industry and manufacturing 20 industry (MLCC) by using the exploratory spatial data analysis method. Finally, the spatial 21 econometric model is used to analyze the driving factors of the MLCC. The results show: The 22 ecological efficiency of the manufacturing industry has steadily improved. The ecological efficiency 23 of the logistics industry presents the rising trend in fluctuation. The level of the coupling coordination 24 development between the logistics and manufacturing industries is high. The results of the spatial 25 heterogeneity analysis show that the spatial differentiation of high-high agglomeration and low-low 26 agglomeration is obvious. The spatial agglomeration characteristics are relatively stable, and the 27 spatial diffusion effect is strong; In space, the MLCC shows a trend of developing from multiple 28 agglomeration areas to one agglomeration area. The results of driving factor analysis show that 29 foreign direct investment(FDI), government intervention(GI) and human capital(HP) have positive 30 effects on the MLCC, while industrial structure(IS), environmental regulation(ER) and energy 31 intensity(EI) have negative effects on the MLCC. At present, with the rapid development of the social economy, China's logistics demand is increasing 37 rapidly, and the scale of logistics is expanding rapidly. How to continuously improve the quality of 38 the logistics industry while reducing energy consumption and carbon dioxide emissions is an urgent 39 problem to be considered. Similarly, while China's manufacturing industry has made great 40 achievements, it has also encountered various environmental problems in the process of 41 industrialization similar to developed countries in recent centuries. The extensive economic growth 42 mode of high input and high-energy consumption leads to the problems of low-resource utilization 43 efficiency and the aggravation of environmental pollution (De Koster,2003). 44 The logistics and manufacturing industries have a naturally close relationship; they influence each 45 other and develop interactively(W.L. . Under the background of a new round of 46 technological innovation and industrial reform, logistics industry and manufacturing industry are 47 undergoing the transformation from traditional industry to digital industry (Kaur et al., 2017). The 48 coupling coordination of the logistics and manufacturing industries determines the sustainable 49 development level of the "two industries" and the comprehensive competitiveness of the regional 50 economy. It is conducive to the sustainable development of logistics and manufacturing and the 51 stability of sustainable supply chain. In recent years, the development trend of coupling coordination 52 and integration between China's logistics and manufacturing industries has increased, but the degree 53 of integration is not high enough, the scope is not wide enough, and the degree is not deep enough, 54 which does not meet the general requirements of promoting the formation of a strong domestic 55 market, building a modern economic system, and adapting to the new pattern of "double cycle" 56 development(Notice,2020). 57 According to the statistics of IEA database, the global power and heat production industry 58 contributes 42% of carbon dioxide emissions, and the industry (including manufacturing, 59 construction and non-fuel mining) and transportation industry contribute 18.4% and 24.6% 60 respectively. In China, industry and transportation contributed 27.9% and 9.7% respectively. In 2020, 61 China put forward the goal of striving to reach the peak of carbon dioxide emission by 2030 and 62 achieving carbon neutralization by 2060(Hu,2021). At the same time, China's economy has entered 63 the stage of high-quality development, and the cultivation of industrial clusters with international 64 competitiveness has become a new power source to achieve high-quality economic development. In 65 2018, the integrated development of the Yangtze River Delta became a national strategy. The 66 Yangtze River Delta is an important gathering area for China's advanced manufacturing industry. 67 Through informatization, networking and the integration effect of modern logistics, the logistics cost 68 of manufacturing enterprises can be effectively reduced, and more resources can be concentrated to 69 develop the core competitiveness of the manufacturing industry. Therefore, it is of great practical 70 significance to study the development of the coupling coordination between the logistics and 71 manufacturing industries to promote the transformation and upgrading of the manufacturing industry 72 in the Yangtze River Delta region, to achieve the goal of "made in China 2025" and to realize the 73 high-quality development of the regional economy. It is further conducive to achieving the goal of 74 reaching the carbon peak in 2030. 75 Since the concept of ecological benefit was first proposed by schartger and Sturm (1990) the logistics industry and manufacturing industry has not been explored yet. Therefore, the current 105 article aims to determine the ecological efficiency of the coupling coordination development of 106 logistics industry and manufacturing industry considering environmental effects, as well as the 107 spatial model (spatial heterogeneity, spatial agglomeration and driver analysis) and trend dynamic 108 evolution of the coupling coordination development efficiency of the two industries. 109 The potential academic contributions of this paper are as follows: (1) The coupling coordination 110 measurement system of logistics industry and manufacturing industry considering carbon emission is 111 established.
(2) This paper analyzes the spatial effect of the coupling coordination degree between 112 logistics industry and manufacturing industry by using spatial correlation model, and discusses the 113 influencing factors of coupling coordination by using spatial econometric model. The analysis has 114 certain policy guidance and theoretical significance. The rest of this study is as follows. The second 115 part presents introduces the relevant models. The third part explains the indicators and unexpected 116 data. The fourth part presents the empirical analysis. Finally, this paper summarizes the conclusions 117 and discusses relevant suggestions. The coupling coordination Coupling degree indicates the degree of mutual influence between two 121 or more systems or elements due to interaction. Coupling coordination degree is the horizontal 122 embodiment of mutual promotion and benign interaction between systems. In recent years, the theory 123 of coupling coordination has been widely used in many fields of transportation, such as control 124 operation strategy, planning strategy, predictive control and so on. In the composite system 125 composed of logistics industry and manufacturing industry, the coordinated development of two 126 industries refers to the evolution of two subsystems and their components towards a harmonious and 127 consistent direction, and the coupling of two industries refers to the mutual influence and interaction 128 between two subsystems and their components. The order parameters in the two-industry complex 129 system are the fundamental variables that determine the evolution trend of the system. The key to the 130 evolution of the two-industry complex system is the synergy between the internal order parameters, 131 and the coupling coordination degree is the measure reflecting the synergy. 132 Where, C represents the coupling degree; U1 and U2 represent the comprehensive order 134 parameters of logistics and manufacturing respectively, expressed by logistics efficiency and 135 manufacturing efficiency under low-carbon constraints. 136 The coupling degree C cannot reflect the overall level of logistics efficiency and manufacturing 137 efficiency and the synergy between them. For example, when the logistics efficiency and 138 manufacturing efficiency are very low, the coupling degree will be very high. Therefore, based on 139 relevant research results, a coupling coordination degree model is constructed: 140 12 , (4) Where, D represents coupling coordination degree; C represents coupling degree; T represents the 142 comprehensive coordination index of logistics efficiency and manufacturing efficiency under low-143 carbon constraints; α and β are undetermined parameters, and are assigned 0.5 respectively according 144 to the practices of most scholars. The coupling coordination degree is between 0 and 1, and the 145 median segmentation method is adopted to divide the coupling coordination degree into five grades 146 (Table 1). 147 In order to give consideration to the generality and rationality of the index system, this paper selects 151 the comprehensive technical efficiency as the decision-making basis for evaluating the 152 comprehensive order parameters of logistics industry and manufacturing industry. 153 In view of the heterogeneity of production technology in the Yangtze River Delta, this paper uses the 154 common frontier model based on SBM method to measure the comprehensive technical efficiency of 155 manufacturing and logistics industry. The calculation is realized with MaxDEA ultra7.0 software. 156

Spatial diffusion effect model 157 1) Global spatial correlation 158
Global spatial correlation is used to analyze the spatial association and spatial difference of the 159 research object and to determine whether there is spatial diffusion. It is generally measured by the 160 global Moran index (Moran,1953). 161 where I is the Moran index and, According to the spatial distribution, the expected value of standardized Moran's I was calculated 169 The variance of normal distribution is as follows： Local spatial autocorrelation is the local expression of Moran index, which is the test form of 182 agglomeration and dispersion effect of local area. It reflects the correlation degree between the 183 attribute value of a region and the attribute value of adjacent region. 184 Where: Ii is the local Moran index of area I, and other parameters are the same as above. 186 If Ii is greater than 0, it means that the local area and the surrounding area present similar spatial 187 agglomeration (high high or low low agglomeration). If Ii is less than 0, it shows different spatial 188 agglomeration (high low or low high agglomeration). The cluster of regions can be directly analyzed 189 by Moran scatter diagram, and the local Moran index is still tested by Z statistic. the error terms of the model, then using the SEM, which can be expressed as follows: 194 Z indicates explained variables and is the vector of (n×1); X indicates explanatory variables, α 196 indicates the regression coefficient, and is the vector of (k×1); μ indicates the random error vector; δ 197 is the coefficient of spatial correlation between regression residuals; W indicates the spatial weight 198 matrix (n×n); and ε indicates distributed random terms independently. If the model has significant 199 dependence between explained variables and would affect its results, using SLM model, which can 200 be expressed as follows: 201 ρ indicates the coefficient of endogenous interaction effects (WY), with its size representing the 203 degree of spatial diffusion and spatial dependence. A significant value shows that definite spatial 204 dependence exists in explained variables. If the model considers both the endogenous interaction 205 effect of the error term and the endogenous interaction effect of explained variables, then using SDM, 206 which can be expressed as follows: 207 θ is the spillover coefficient, others are as the same with SLM model. 209 210

Selection of indicators 212
With reference to most very weak research on the coupling coordination efficiency of these two 213 industries, combined with some selection principles (significance, practicability, data availability, 214 isotropy, etc.) of the indicator system, the manufacturing-and-logistics-industry, coupling 215 coordination-development, efficiency-evaluation, index system is constructed from the perspective of Environmental Statistical Yearbook, the China Energy Statistical Yearbook and the school library. 220 Due to the availability of data, the transportation industry and warehousing, postal and 221 telecommunications industrial data are generally used as logistics industrial data (since China has not 222 yet established a relatively complete statistical system for the logistics industry, and the statistical 223 caliber of the logistics industry in each country is not the same, the model and index selection in this 224 paper are only applicable to a single country or region with unified logistics industrial data statistics). 225 The specific selection indicators are as follows. 226  The coupling coordination efficiency of logistics industry and manufacturing industry is the 230 comparative relationship between input and output or between cost and income in the process of 231 business activities of the whole system formed by logistics industry and manufacturing industry. 232 Therefore, the ecological efficiency of the coupling coordination between the manufacturing industry 233 and the logistics industry means that the ecological efficiency of the logistics industry and the 234 manufacturing industry are input factors to each other, which has an impact on the system. Therefore, 235 refer to the research methods of Wang Zhenzhen (2017), on the basis of Table 2 and table 3, after  236 considering the impact of both sides on each other, the index system as shown in Table 4 is  237 constructed (Wang Zhenzhen et al.,2017). 238  The added value of the secondary industry as the proportion of the regional GDP % Government Intervention (GI) The proportion of public finance expenditure to regional GDP (1)the concept of desirable 0utputs and undesirable 0utputs 243 In the production process of general economic system, we always hope that the smaller the input is, 244 the better the output is. That is to say, we can produce as many outputs as possible with the smallest 245 input, and the relative efficiency of production unit is also higher. The core idea of traditional DEA 246 model in dealing with input-output in efficiency evaluation is also based on this. The output 247 mentioned here is the expected output, which is the main purpose of economic production, and refers 248 to the output with benefit. However, the production process will inevitably produce some negative 249 effects on economic development, people do not want to produce output. The output mentioned here 250 is undesirable outputs, that is, the output that people do not expect. It is a subsidiary product of 251 economic production, which means the output that is not beneficial or even harmful. For example, 252 the paper products industry, while producing, emits industrial wastewater, waste gas, waste and other 253 pollutants. The paper or paper products here are called expected output, and industrial wastewater, 254 waste gas, waste and other pollutants are called undesirable outputs. Considering environmental 255 variables and paying attention to undesirable outputs will have important theoretical and practical 256 significance for transforming the mode of economic development, strengthening environmental 257 protection, promoting sustainable development, and establishing a saving and efficient industry. 258 (2)the undesirable 0utputs data of manufacturing industry and logistics industry 259 respectively. The calculation formula is: 269 Where i is the type of fuel, Ei is the consumption of i fuels, CFi is the calorific value of i fuels, CCi is 271 the carbon content of i fuels, and COFi is the oxidation factor of the fuel. value are obtained (see Table 6). The results show that the index of the logistics and manufacturing 300 industries of 25 cities in the Yangtze River Delta is positively correlated. The Z value for each year is 301 positive, and the p-value is less than α (0.05), which indicates that the spatial autocorrelation is 302 significant. This indicates that the regions with similar development levels (high or low) of the 303 logistics and manufacturing industries in the Yangtze River Delta are concentrated in space; that is, 304 the cities with higher coupling coordination levels of the two industries have higher coupling 305 coordination levels of the two industries in their peripheral cities, and vice versa. From 2006 to 2019, 306 the Moran index of the logistics and manufacturing industry in the Yangtze River Delta showed a 307 trend of first rising and then declining, but the Z value of the coupling coordination efficiency of the 308 two industries was greater than 5, indicating that the spatial distribution of the logistics and 309 manufacturing industries in the Yangtze River Delta during this period showed a globally significant 310 similar level of agglomeration characteristics. 311

2) Empirical analysis of local spatial autocorrelation 314
In order to reflect the spatial structure changes of logistics industry and manufacturing industry in the 315 Yangtze River Delta region from 2006 to 2019, according to the local spatial autocorrelation analysis 316 method, this paper selects the coupling coordination efficiency of logistics industry and 317 manufacturing industry in 2006, 2013 and 2019 as the research data, and analyzes the spatial 318 characteristics and changes of the coupling coordination efficiency between each city and adjacent 319 cities in the Yangtze River Delta region. The results ( Table 7) show that there is an obvious spatial 320 distribution pattern of logistics and manufacturing industry in the Yangtze River Delta. 321 According to the results of Lisa, each city is divided into four quadrants according to its nature. As 324 shown in Table 7, this paper selects several representative cities for analysis. 325 The first quadrant is the HH agglomeration type of logistics and manufacturing coupling the logistics and manufacturing coupling coordination efficiency of the city is higher, and the local 366 Moran values of the surrounding cities are relatively low, that is, the cities with high logistics and 367 manufacturing coupling coordination efficiency are ignored Surrounded by low value cities, it has 368 spatial spillover effect on the surrounding cities. It can be seen from Table 7  To sum up, the spatial agglomeration attribute of logistics and manufacturing coupling coordination 376 efficiency in Yangtze River Delta is relatively stable, and the spatial diffusion effect is strong. In 377 recent years, the rise of Suzhou, Wuxi and Changzhou makes the Yangtze River delta form a 378 development trend from multiple agglomeration areas to relatively concentrated agglomeration areas. 379 Jiaxing and Taizhou are the typical cities in the Low-High agglomeration type. These cities are in the 380 transition zone of Shanghai, Nanjing and Hangzhou megalopolis, surrounded by these developed 381 cities, and are the transition zone of the study area megalopolis for factor flow. Lishui and Suqian are 382 typical cities in Low-Low cluster type. The development level of logistics industry in these cities is 383 backward, the foundation of manufacturing industry is poor, and the coupling coordination efficiency 384 of the two industries in the surrounding areas is also low, which is easy to form marginal areas. The 385 typical city in High-Low agglomeration type is Wenzhou, where the export-oriented economy is 386 developed and the manufacturing industry has a good foundation, resulting in spatial spillover 387 effect (Jiang P et

Empirical analysis of spatial econometrics 389
According to the previous analysis, the spatial autocorrelation test of coupling coordination 390 efficiency of logistics industry and manufacturing industry indicates that it exists a spatial 391 agglomeration. To find a proper spatial econometric model, this paper use (robust) lagrange 392 multipliers (LM) test, the likelihood ratio (LR) test, the Wald test and Hausman test ( Table 8). 393  LM test results show that the spatial lag effect is more significant than the spatial error effect. 397 Therefore, it is more suitable to use spatial Durbin model. 398 It is noteworthy that the coefficients of the SDM model do not directly reflect the marginal effects of 399 the corresponding explanatory variables on the dependent variable (LeSage and Pace, 2010), so it 400 needs to verify spatial effects from the total effect, the direct effects and indirect effect ( Table 9). ***Significance level at 1%. **Significance level at 5%. *Significance level at 10%. 403 The regression results show in Table 8, the industrial structure (IS) has a negative impact on the 404 coupling coordination ecological efficiency, it fully shows that the increase in the proportion of 405 China's logistics industry and manufacturing industry has a negative effect on ecological efficiency. 406 With the acceleration of industrialization, the secondary industry drives economic growth will not 407 change in a short period of time. are that the degree of marketization is not enough, the direct information asymmetry of the two 430 industries, and the willingness of active cooperation is still not high. Therefore, government 431 intervention can play a positive role in standardizing the market order, coordinating the operation of 432 logistics and manufacturing industries, continuously guiding the two industries to actively cooperate, 433 and improving the coupling coordination efficiency of the two industries. 434 The energy intensity (EI) direct coefficient is -0.028, indicating that the energy intensity has a 435 significant negative effect on the improvement of logistics ecological efficiency. The results of the local spatial autocorrelation analysis show that: The coupling coordination 474 ecological efficiency of the two industries in the Yangtze River Delta presents a development trend 475 from multiple agglomeration areas to one agglomeration area in space, which indicates that the 476 coupling coordination of logistics industry and manufacturing industry in the Yangtze River Delta 477 has entered a relatively centralized development stage. 478 The results of empirical analysis of spatial econometrics show that FDI, GI and HC have positive  479  effects on the coupling coordination ecological efficiency of the two industries, while IS, ER and EI  480 have negative effects on the coupling coordination ecological efficiency of the two industries. 481

6.SUGGESTIONS AND POLICY IMPLICATIONS 482
According to the above research conclusion, this paper presents some suggestions for the coupling 483 coordination development of the logistics and manufacturing industries in the Yangtze River Delta 484 region. 485 First, although the efficiency of the regional logistics industry is at a high level, this still cannot meet 486 the needs of the manufacturing industry, and the coupling coordination ecological efficiency of the 487 two industries lags behind the efficiency of the manufacturing industry. Therefore, in order to 488 improve the level of coupling coordination development, it needs to continue to make greater effort 489 to develop high-end logistics dominated by high-tech; which can greatly improve the service level, 490 specialization level and economies of scale that are advantageous for logistics enterprises; and to help 491 the manufacturing industry to reduce costs and to improve efficiency. 492 Second, the transformation and upgrading of the local manufacturing industry needs to continue to be 493 promoted. Although the efficiency of the manufacturing industry in the research area is at a high 494 level, there is still room for development. In the future, the transformation and upgrading of China's 495 manufacturing industry should be promoted under the guidance of the Internet of Things, 496 Blockchain, artificial intelligence, big data, cloud computing and other advanced technologies to 497 achieve the goal of "made in China 2025". 498 Third, it is increasing concern to society about the environment impact of the development of the 499 logistics and manufacturing industries. The development policies of the logistics and manufacturing 500 industries formulated by relevant managers should not only consider the economic benefits but also 501 consider the impact on the environment. 502 Fourth, energy conservation and emission-reduction technology should be promoted. In order to 503 achieve the goal of reaching the carbon peak by 2030, the government should increase investment in 504 the development of energy-saving and emission-reduction technology, enhance the exchange of 505 advanced technology and the management experience of energy savings and environmental 506 protection, actively develop and promote energy savings and emission-reduction technology to 507 effectively improve the increasingly serious problem of environmental pollution. 508 Fifth, the government should actively develop the circular economy and realize the recycling of 509 resources and sustainable development of logistics and manufacturing. Government managers can 510 expand the scope of circular economy pilots to spread all over the country, formulate environmental 511 regulation policies reasonably, encourage industries with high-energy consumption and large 512 amounts of pollution to develop an internal circular economy, and reduce undesirable outputs. 513 The limitations of this paper include the following. Regarding research data acquisition: because the 514 logistics industry is a network industry involved in many other industries, currently, a systematic 515 evaluation index of the efficiency of the logistics industry is still missing; there are currently no 516 authoritative logistics statistical indicators in China; and compared to urban logistics and 517 manufacturing data, the availability of undesirable outputs data is poorer. 518 Regarding future research: with the development of informatization, big data and the Internet of 519 Things, more extensive research can be conducted on index selection and data acquisition in the 520 future. This study is limited to the Yangtze River Delta region. The interaction between 'the logistics 521 and manufacturing industries has a different development status in different regions in China. In the 522 future, the study area can be extended to the entire country. 523 Ethical Approval and consent to participate: Not applicable 524

Consent for publication: Not applicable 525
Declarations 526 No potential conflict of interest was reported by the author(s). 527

Data availability 528
The original contributions presented in the study are included in the article/Supplementary Material, 529 further inquiries can be directed to the corresponding author.