Campus Secondhand Book Swap Station Regression Analysis of Resource Utilization

A large number of discarded academic books will be left on the university campus after the end of the semester. Observational data from three local comprehensive universities show a large amount of textbooks have been discarded. At these universities, the end-of-semester clean-up has left an average of 3.2 tons of unused textbooks, reference books and course materials that have been placed in campus waste bins or transferred to unprocessed bulk recycling stations at the end of each semester. The above linear disposal path is a severe waste of resources. Functionally, these materials still have good structure and teaching value; however, they are being scrapped prematurely in the life cycle of use before the necessary operation steps for a circular resource system can be realised, thus reducing the possibility of reusing them.

At the same time, there are an increasing number of low-carbon circular resource solutions promoted by institutions and students. In terms of operation, the campus sustainability plan aims to reduce new paper consumption and lower the overall carbon emissions of the university's teaching activities. The production of replacement academic materials will require large-scale wood extraction and extensive energy-intensive processing and logistics distribution, resulting in high greenhouse gas emissions. The circular resource model of systematic book swapping can directly reduce the environmental cost of the closed-loop institutional economy and meet the demand for low-carbon circular resources.

Based on the collected data from the universities, it has been found that the current ways of disposing of campus circular resources do not meet the expected goals of the campaign. Although there are many reusable academic texts in quantity and there is a demand for green alternatives in nature, no structured or regularised exchange system has been established to address the problem of waste. Students have materials that still have some value in use, but due to the high costs associated with finding other people who want them in an unsystematised manner, they usually throw these away. This is a clear mismatch, that is to say, waste is generated in a linear way rather than being recycled in a closed loop; thus, it is the foundation of the research. Quantifying this difference with regression analysis will help us understand the causes of inefficient use, and then we can develop a plan for better circulation of campus resources.

The main purpose of the study is to systematically discover the main reasons that have a significant impact on the degree of resource use at the campus secondhand book swap stations. Based on the construction of a strict theoretical model, the study will separate the variables of station accessibility, diversity of inventory, and platform usability to determine their basic definitions and operation paths in the campus environment. Secondly, this paper will use quantitative regression analysis to determine how much each of the relevant factors influences it. Apply some regression models to figure out the rules of change in the above variables and their impact on the usage rate accurately; therefore, this analysis will now be based on empirical data rather than just a qualitative description. Thirdly, based on the empirical results, some measures for optimization at the stations will be put forward. The above construction path of the implementation scheme should improve the operational efficiency of the organisation and accelerate the turnover of resources through the adjustment of management methods according to statistical laws.

The value of this research is due to the contribution of practice and theory. In practice, it can help reduce the cost of textbooks for students by means of a mechanism that has been tested and validated, thus promoting fair education. Optimise the operations of the swap stations and extend the life of campus culture resources, complying with the rules on resource saving of the institution and industry. The optimal model derived can also provide other campus idle resource sharing projects with a new idea for managing various types of idle assets. For academic purposes, it can provide some examples of the research on the distribution of second-hand cultural goods on campus. At present, most studies only qualitatively describe the behaviour of students in sharing, and this paper contributes a solid, statistically verified model to explain the changes in the structure of resource use, thus strengthening the methodological quality and theoretical completeness of the field.

The research scope of this paper is the second-hand book exchange stations in some Chinese undergraduate universities that do not aim to make a profit and are not involved in business; they have been operating continuously for at least eight months in one academic year. Therefore, we will systematically exclude all online second-hand book trading platforms, as they have different ways of distributing digital resources and different logistical systems that do not fit the physical space constraints. In addition, due to the profit motive of the second-hand bookstores off campus, they have been excluded to remove the interference of this confounding factor and maintain the purpose of studying how resources circulate among peers in the academic community. To ensure that the selected stations have operated for at least eight months and have gathered sufficient stable data over time for reliable regression analysis, rather than only temporary or seasonal deviations, this requirement has been set.

The main problem studied under this limited range is to determine the direction and strength of the relationship between several possible factors and the actual use efficiency of these swap stations. At present, the operation of most other campus facilities is based on empirical observation and heuristic adjustment, resulting in a vague operational state based on experience that cannot systematically optimise the turnover of resources. To convert the operational strategy from subjective intuition to evidence-based regression modelling, this paper will measure how changes in certain variables (such as spatial layout, accessibility, frequency of updates in collections, and institutional policy support) affect the utilisation rate in statistics. To be able to rationally distribute the space and funds of universities by knowing what and how much of these factors are affected, as well as in which direction, and thus to maximise the ecological and educational benefits of non-profit book circulation on campus, we must first identify the exact magnitude and direction of influence of these factors. Therefore, this paper can basically solve the problem of the lack of a method for quantitative evaluation of campus resource sharing mechanisms.

The overall research framework of this thesis is built on a strict logic chain, moving from problem conceptualisation to empirical verification and finally to strategy. First, we need to know that the campus second-hand book exchange stations have a low utilisation rate of resources; then, from the old document about resource circulation and regression analysis... Next, the model is used to define explanatory and response variables, and then the econometric model is run to conduct regression analysis. After some tests on the model's fit and the size of the coefficient, the results of the test are used to give some suggestions for improving the model.

Based on the logic of the chapter arrangement, each subsequent chapter is designed in detail to help achieve the main purpose of this research: to quantify the reasons for resource utilisation and present evidence-based suggestions for improvement. Chapter One lays the groundwork, clarifying the background, significance and research methods of the study to support the following empirical analysis. Chapter Two will systematically sort out the theoretical basis and related research, point out what is lacking in previous studies and put forward hypotheses that guide the selection of variables in the model. Chapter Three is about the Research Design; it explains how to operate variables, collect data and build a Multiple Regression Model in line with the conditions of the Classical Linear Regression Model. Chapter Four will perform an empirical analysis, use a regression model to estimate, test the stability of the results, and explain the economic and statistical significance of the coefficients to verify the research hypotheses. Finally, based on the above findings, Chapter Five will propose targeted improvement plans within the scope of campus operations, but also recognise that this study has some limitations and indicate directions for further research in the future. It is arranged in a chronological order, so the structure is good, the research method is clear, and there is logical development.

Because the semantics of this paper are not clear, we need to define it further. Previous studies often regard these stations as commercial second-hand bookshops or casual online exchange groups; therefore, they have not shown a clear institutional logic. Based on the operation characteristics of the sample stations in this study, the campus second-hand book swap station is a non-profit or low-profit public service facility on campus. Its main purpose is to provide a free or low-cost exchange and circulation platform for spare textbooks, teaching reference books, and other cultural books among the campus population. This definition emphasises a key difference from commercial enterprises: while commercial bookshops operate on the principle of arbitrage to increase profit by buying and selling books, swap stations are based on the logic of resource circulation to improve the utilisation rate of idle academic resources and reduce the financial pressure on students, and they are subject to the laws and regulations of universities. In addition, this idea is different from pure online book exchange groups. While the latter uses the Internet to match people and provides a variety of books for free, it cannot offer a variety of places to read. Swap stations have a specific space, classify and place books uniformly, and provide convenient services for borrowing and exchanging books; thus, they are more reliable and transparent. The process is usually that users hand over their idle books in a standardised category, and then the facility organises the collected materials and distributes them to other users according to a point system or direct exchange. Through the institutionalisation of knowledge resource transfer, these stations are no longer limited by the temporary nature of online meetings or by the profit motives of business; they have become an important part of the campus circular economy and help to build a sustainable, community-oriented learning environment. Based on this accurate conceptual boundary, the next step is to conduct regression analysis on the utilisation efficiency of resources.

There are three related reasons for the circular use of campus cultural goods. Firstly, the circular economy theory can be applied to the small-scale environment of the campus, change the old-fashioned linear mode of acquisition and disposal, and build a closed-loop system of reuse and distribution. Second-hand books are cheap to buy and in good condition in a college campus, so they are ideal for recycling. The above theory can provide the structure for how swap stations stop waste streams; thus, the selection of evaluation indicators such as circulation rate and retention rate should be made based on them to measure the deviation from linear consumption. The second is to extend the life cycle of cultural resources through this theory. Textbooks have a long life; they can be used in more than one semester and are relatively durable. Through the swap mechanism, the working process of extending the use time of these items and delaying their terminal obsolescence will be promoted. This theoretical basis can be used to select independent variables for the regression model later, and the frequency of exchange interventions and the rate of physical deterioration will be used to measure how life-cycle extension has affected total resource use. Third, in terms of campus public resource sharing, user-perceived value theory is used to study the reasons behind people's participation in this circular system. The success of the swap station is determined by whether students believe it is convenient and beneficial to use it. If the perceived economic benefits and procedural efficiency are higher than the perceived costs, there will be an increase in participation. The above theory will provide a behavioural basis for explaining the results of the regression, why some structural and spatial variables have high coefficients; they directly affect users' perception of value and thus their level of use. Combining the three theories can provide a good logical framework for this paper; the circular economy is the system boundary, life cycle extension is the measurable physical mechanism, and the perceived value theory explains the micro-behavioural reasons. Together, they form a new theoretical matrix that explains the choice of variables and ensures that the empirical regression analysis of swap station resource utilisation has a sound structure and good ideas.

The research on the efficiency of secondhand textbook circulation abroad, particularly in North America and Northern Europe in the last ten years, has been relatively well-developed with quantitative ideas and operating methods. In North America, the general model is that university textbook sharing is carried out through a centralised campus bookstore or a distributed digital platform, and it is used to buy back and rent textbooks. Regression analysis and discrete choice models have been widely used to study the influencing factors of the reuse rate. The important reasons for this are that the location of the swap is nearby at any time, the price decline curve is relatively large compared to that of a new version, and the curriculum is stable. Based on the above results, a one-standard-deviation increase in the proximity of accessibility is associated with a large increase in the circulation probability; therefore, to enhance the circulation rate, it is necessary to reduce barriers to transactions.

Sustainable circulation models in Northern Europe are led by student organisations and library officials, and they are combined with the general idea of the circular economy to reuse textbooks. It is found that conditional quality grading systems play an important role in reducing information asymmetry and can increase transaction trust and subsequent reuse rates. Incentive schemes differ according to whether they are monetary discounts or altruistic/sustainability nudges in other countries. Regression discontinuity design shows that the direct financial buyback guarantee can increase the initial user's participation, and eco-centric behavioural incentive can promote long-term retention and cross-period circulation to a certain extent. The above mature Quantitative Analysis Framework of transaction data is provided with a good way to calculate the circulation efficiency.

But there are some differences in the relevant context of overseas campus environments and domestic university situations. The overseas model is based on a highly distributed textbook selection system; thus, different textbooks will be used for different teachers, leading to uncertain demand and frequent edition changes that require certain dynamic pricing rules. However, in China, it is generally based on unified curriculum standards and consolidated lists of textbooks; as a result, demand is relatively homogeneous and stable. Therefore, the above differences show that domestic swap stations can make predictions on the circulation inventory more accurately at the beginning of the semester; however, they need to adjust their regression models based on the domestic institutional rules and campus space management policies of China, rather than directly copy the overseas market-driven operation model.

The development of the campus idle resource-sharing system for college students has recently been the focus of many studies in China, especially for second-hand book swap stations. According to some materials, it is known that these sharing projects are currently in the operation stage of peer-to-peer transfer, centralised warehouse and digital platform matching. Generally speaking, the above mechanisms are conducive to the circulation of academic materials and help reduce the financial pressure on students. The problems of operation are mentioned in a large number of works of literature. Trust among the members of the group is not good; it is not possible to obtain timely information about books; there is insufficient institutional support according to university rules, and there is a lack of stable operating funds over a long period of time.

Based on the above problems, the general proposal of optimisation ideas in China now is to build a standardised credit evaluation system, improve platform-based identity verification methods, and promote organised cooperation between student organisations and university administrative departments for stable venues and operating guidelines. Although there has been a great deal of discussion on the above, there is still a problem of method in the current domestic research. Most of the existing studies are in the stage of qualitative description and mechanism discussion, and they often use case studies, theoretical construction, and descriptive surveys to illustrate the concept of sharing mechanisms and its social significance.

Therefore, there is very little in-depth research on the actual effect of resource utilisation in swap stations. In recent years, there have been very few domestic studies that have employed rigorous econometric models or regression analysis to objectively determine the factors affecting swap efficiency, to quantify the marginal effects of various incentive mechanisms on circulation rates, or to statistically assess the long-term sustainability of resource use. Therefore, the above research has not been able to determine precisely how idle resources on campus are shared and what results they have; thus, it is necessary to conduct some quantifiable empirical studies to provide a basis for improving the system.

A systematic comparison of the literature at home and abroad shows that there are differences and some problems. Most of the research on circular economy in foreign countries has been conducted on a large scale, and scholars have used more complex econometric methods to study digital sharing platforms and the general second-hand market. Most of the domestic research on campus sustainability is descriptive in nature, and it has been carried out by universities in recent years; in particular, studies of campus sustainability are relatively scarce and have employed relatively few inference models. Both sides recognise the ecological and economic value of book recirculation, but there is no targeted regression quantitative analysis to study the resource utilisation efficiency of a physical campus secondhand book swap station. The current quantitative models are not suitable for the specific spatial limits, peer-to-peer interaction, and institutional policies of the physical campus environment, so the reasons for the low usage rate of the swap station are not yet clear. In order to fill this gap, this paper will use some mature theoretical models and research methods from previous studies. Based on the Theory of Planned Behaviour, we will determine the reasons for students' participation in resource allocation, and from the perspective of macro-operational factors, we will analyse spatial accessibility and acquisition frequency. In terms of constructing a regression model, this paper will refer to the econometric methods used in generalised second-hand market research and change the functional forms to adapt to the limited space and population characteristics of a university campus. A multiple linear regression model will be used to explore the interaction effects of physical proximity and academic time. For the purpose of interpreting the results, according to the theory of marginal analysis in previous ecological economics studies, not only will the estimated coefficients be statistically tested but also the practical significance of their marginal impact on the overall use of resources will be considered. Based on this strong theoretical basis, the subsequent empirical estimation will also be rigorous in method and structurally consistent with the specific mechanism of resource circulation on campus.

In the regression model of resource utilisation for the Campus Second-hand Book Swap Station, the strict definition and quantitative measurement of the dependent variable are essential for the analytical framework to be reasonable. The dependent variable in this study is the second-hand book reuse rate of the swap station in a year. Essentially, this rate is the ratio of all the books that have completed an effective exchange or transfer through the swap station in one year to the total number of valid second-hand books that have been added to the inventory of the swap station in the same period. To sum up, we have made the abstract idea of resource use a specific, continuous numerical index that fits ordinary least-squares regression estimation very well.

The basic idea behind the measurement is to directly show how efficient the platform is in promoting a circular economy in the campus environment. In other words, the numerator is the number of successful transactions, and the purpose of the swap station is to turn idle academic resources into active ones, not just accumulate them. Therefore, the denominator (Total number of valid inventories) needs to be the actual amount of transactable resources. So that, books that are damaged and cannot be repaired and then scrapped are not counted in the total inventory. Excluding them is an important change in the method; items that are beyond repair have no more residual value for educational use, so they should not be artificially increased in the denominator, which will regularly reduce the reuse rate and distort the actual use efficiency of the operation system.

Based on the above exclusion rules, the change in the number of people participating in the second-hand book swap is not directly affected by this policy. It can be distinguished from the other reasons for the low number of people. Based on the data analysis results, it can be used to compare the operating status of the campus library system with the expected goals of the campus's green development; this will be the basis for future investigation.

A group of independent variables has been organised in three categories of the operation, user and facility to be used to explain the reasons for the improvement in resource utilisation efficiency. The operational factor dimension is the degree of management and procedure at the swap station. There are three main variables here, and the total number of systematic categorisation and culling activities per year is used to calculate how many activities are organised by the station; the average number of new second-hand books added to the bookshelf each month; the reward mechanism for users to donate books and so on; this is counted on a discrete ordinal scale with zero meaning no incentive, one meaning non-monetary credit and two representing a tangible reward redemption system. The above factors are the foundation of the idea; thus, continuous operation will not cause inventory issues and directly drive circulation.

The user factor dimension is the demand-side situation and space distance that affect people's behaviour. It is the proportion of students on campus who have heard about the swap station and are thus informed of the service; it has been calculated as the percentage of all students aware of the service. The average distance between the student dormitory and the swap station is given in metres to indicate the space cost; the mean of all student satisfaction with book types is obtained based on a Likert scale survey of 1-5. Since the use of resources is closely related to people's will, these variables can be used to examine how well a service is known about, how convenient it is in space and how suitable the stock of goods is for individual needs.

The level of the construction is also a factor that affects the service supply of the facilities, and it needs to be considered. It includes the size of the station's available storage area, measured in square metres, to show the size limit of the space; the total operating time per week; and whether there is an automatic circulation machine, that is, a binary variable where one is present and zero is absent. The above factors are the theoretical maximum throughput and operating elasticity of the station, and they determine the limits of resource use in terms of efficiency. By strictly setting the above quantifiable criteria, the independent variable group can ensure the empirical nature and repeatability of the regression model in the later stage.

The sample data of this study comes from a field survey of 32 physical campus secondhand book swap stations at 11 different universities, together with the daily operation log data of these stations obtained by working with the campus student associations responsible for them. The cross-sectional survey data has structural characteristics and resource inputs, and the longitudinal logs provide dynamic use indicators. The design of the multi-source will increase the flexibility of the regression model for resource use.

To ensure the basic data quality for the subsequent regression analysis, a strict data preprocessing process will be conducted. First, identify and eliminate abnormal values through the method of interquartile range. If there is any observation $x_i$ that does not meet the condition of the first and third quartiles, that is, if $x_i < Q_1 - 1.5 \times IQR$ or $x_i > Q_3 + 1.5 \times IQR$, it will be removed as an outlier. Here, $Q_1$ is the first quartile and $Q_3$ is the third quartile, and $IQR = Q_3 - Q_1$. Second, multiple linear imputation is used to fill in missing data to keep the sample size and distribution as unchanged as possible and avoid the bias caused by listwise deletion. Third, different indicators will be standardised using Z-score normalisation to guarantee that data from various units can be compared. The transformation is as follows:

$\bar{x}$ is the sample average and $s$ is the standard deviation of the sample.

Finally, the preprocessed sample data must meet the basic prerequisites for ordinary least squares regression. We will check whether the maximum variance inflation factor is below the threshold of 10 to avoid severe multicollinearity. The augmented dataset still has sufficient data points compared with the number of predictors, so it can be used to estimate the unbiased parameters. Therefore, after the above systematic process, the preprocessed data has met the basic data quality requirements for the subsequent regression analysis, and an empirical foundation for academic rigour and reliability has been laid.

Based on the above, the first hypothesis is that the number of book-sorting activities will have a significant positive relationship with the second-hand book reuse rate. The reason for this is that organised curation activities can reduce the difficulty of searching, improve the visibility of goods, and thus drive students to participate and increase the reuse rate.

The second is that the average distance from the student dormitory to the book-borrowing station will be significantly negatively correlated with the second-hand book reuse rate. According to the theory of spatial interaction, swapping behaviour is more sensitive to transaction costs. A longer distance will increase the implicit cost, thus discouraging people from regularly accessing and reducing the reuse of books.

Third, the frequency of online platform updates is assumed to have a significant positive relationship with the second-hand book reuse rate. Higher update frequency can speed up the flow of information, reduce the information gap between book supply and demand, and thus improve matching efficiency and facilitate the physical exchange of books.

Fourth, the degree of book depreciation is assumed to have a significant negative correlation with the second-hand book reuse rate. From the perspective of utility maximisation, severe physical damage has reduced the residual informational and aesthetic utility of a book and will not be attractive to potential recipients, thus decreasing the likelihood of successful exchange.

The above are the first few directional hypotheses.

Set forth the above a priori directional expectation as a theoretical basis for the subsequent empirical test of the regression coefficient. Therefore, this method is based on theoretical guidance and will be more useful in practice to improve the allocation of resources in the station.

A secondhand book's reuse rate can be affected by several factors, so multiple linear regression analysis will be employed to address this problem in this paper. Because the campus exchange station is affected by many factors, such as the number of books in stock, how easy it is to use, and how many promotions are held, it is difficult to identify the reason for changes in the campus exchange station in a single variable. Multiple linear regression is used to separate the influence of each explanatory variable on the explained variable and form a good structural equation with other preset independent variables that are explanatory variables. The regression equation should be tested statistically; otherwise, it will not be considered valid. First, the goodness-of-fit test is based on the coefficient of determination to see how much of the variance in the explained variable can be explained by the explanatory variable in the model and to evaluate the overall explanatory power of the model. Second, to determine whether the regression equation as a whole is statistically significant, an F-test will be conducted; that is to say, the entire model does not have random errors in its prediction. A single-variable significance t-test is required to determine the specific reason why the independent variable is correlated with the reuse rate and guide further variable refinement. Finally, the test for multicollinearity of independent variables is necessary; otherwise, severe linear dependency among the explanatory variables may occur, distorting the coefficient estimation and increasing the standard error. Through the above rules of testing, the established regression model has statistical validity and empirical rationality and can provide a basis for the analysis of resource use patterns.

Descriptive statistical analysis is the first step in empirical research; it can outline the central tendency and dispersion of the preprocessed sample data in a regular way to be used as the basis for regression modelling. We will calculate the minimum, maximum, average and standard deviation of the dependent variable and all independent variables in a strict manner to obtain the quantitative basis for the operation of the campus second-hand book exchange. The minimum and maximum values show the range of actual data collected; thus, it can be seen if there are any obvious outliers or mistakes in the data entry. At the same time, the mean is a measure of central position; it is the average level of operation or resource use in the observed sample. The standard deviation can be used to quantify how spread out the data are around the mean and thus gain some understanding of the differences in the way people use the exchange station and how resources are allocated there.

Histograms and box plots should be used to display the distribution characteristics of each variable to verify the assumptions underlying the subsequent regression analysis. The shape of the data distribution in a histogram shows whether the variable is approximately normally distributed or has skewness and kurtosis, so we will know if data transformation is needed. The box plot shows the interquartile range of the data, and the outliers are located outside the whiskers of the box plot; these are removed to ensure that the data are not unduly affected by the outliers. Based on the above graphs, a simple distribution rule has been derived to see if the resource utilisation data conform to the assumptions of a particular parameter model. In addition, it is possible to observe the degree of overlap and change in the centre of the data of the groups to gain an initial idea of how closely related the independent and dependent variables are to one another. Preliminary analysis is important; it can verify the statistical validity of the sample data and give some empirical direction for specifying the formal regression model, thus ensuring that the subsequent improvement plan for the swap station is based on sound reasons.

The multiple linear regression model for the campus second-hand book swap station resource utilization efficiency is established and executed with the sample data to obtain the specific quantitative results. The overall degree of fit is measured by the R-squared value and adjusted R-squared value. If the R-squared is relatively high, it can be concluded that most of the variance in the dependent variable is explained by the independent variables; and the adjusted R-squared will be adjusted to reduce the explanatory power due to more predictors, so it can show that the model has a good fit. The F-test is used to verify the overall validity of the regression equation, and the F-statistic and its corresponding p-value will be compared with the selected significance level. If the p-value is less than the threshold, then the regression model is statistically significant overall, and the null hypothesis that all the regression coefficients are zero can be rejected.

At the level of individual predictors, the unstandardised regression coefficient is the absolute marginal effect of each independent variable on the swap station's usage rate, assuming that all other variables are fixed. At the same time, the standardised regression coefficient is given to facilitate the comparison of the relative importance of predictors by eliminating the dimensional differences among variables. The t-statistic and the corresponding p-value are used to determine whether a coefficient is statistically significant. If the p-value of an independent variable is below the significance level, it will be retained; otherwise, it will be removed because it does not have a statistically significant correlation with the dependent variable and will be filtered out to reduce noise and prevent overfitting.

Moreover, the multicollinearity test of the Variance Inflation Factor can also be used to test the structural stability of the coefficient estimates. Predictors with a Variance Inflation Factor value less than the critical threshold show that there is no serious multicollinearity, and therefore the results of the significance test are still valid. Based on a comprehensive integration of the above, we can determine that the regression model is reasonably sound and, moreover, find which part of the independent variables has a significant and independent effect on the use of resources to be optimised in the operation.

Based on the calculated standardised regression coefficients, the influencing factors of the resource utilisation of the campus second-hand book exchange station will be analysed in order to determine the actual weights and operational logic of these factors. The standardised coefficients show that the main influencing factors are the volume of student traffic and the variety of books; these have a relatively large weight and play an important role. The notable positive coefficient of student traffic volume is in line with the basic principle of exchange probability: an increase in the number of active users can expand the scope of matches, thus reducing search friction and shortening the circulation period. In addition, the high weight of book categories also shows that inventory heterogeneity is necessary; having a broad range of different disciplines can meet the diverse needs of all students, reduce the number of students who do not want to change books due to substitution, and enhance the matching rate of books.

Based on the above, shelf update frequency and geographical accessibility are the secondary auxiliary influencing factors. They have moderate positive coefficients but do not constitute the main reasons; they serve to complement the operation. The higher the update frequency of the shelf is, the more new goods will be added in time to avoid information asymmetry and inventory backlog, and keep people's perception of new resources high. Geographical accessibility reduces the hidden cost of transactions for participants; however, due to its small weight, it can only be regarded as an auxiliary condition to promote the first-come-first-served competition among students in the city centre's new campus areas and the old campus areas on the other side of the river.

Conversely, the station's static physical area and offline promotional signage are not statistically correlated with the efficiency of resource use in this sample scenario. The lack of statistical significance for physical area shows that the marginal benefit of simply expanding space is small because the amount of inventory does not grow proportionally; instead, it depends on the circulation density of the space. The impact of offline promotional signs is small; in the current digital campus environment, information spread through online academic networks is more dominant, and the traditional way of static visibility cannot effectively promote the desired behaviours in substance. This hierarchical division sets the priority of operation for improving the use of resources at the swap station.

Based on the regression analysis, it can be seen that some factors have a significant impact on how resources are used for the campus secondhand book swap station. Therefore, we should adjust the operating mode of these factors and distribute limited operational resources in a reasonable way to achieve the greatest improvement in the reuse rate. Fundamentally, we need to carry out these changes according to the direction and size of the regression coefficients of the operation, and each strategic change should increase the positive forces or reduce the negative factors found in the statistical model. For example, if the regression results show that the book classification management level has a large positive coefficient, then increasing the level of classification can increase the speed of matching and thus help complete transactions faster. Therefore, the operating mechanism should be adjusted to replace the general alphabetical shelving logic with a multidimensional classification system based on the course categories of each college. In order to increase the positive coefficient, an active notification system will be established to remind students when new books meet the requirements of their current semester, thereby converting the current situation of passive resource provision to an active one for students to use. Otherwise, if the regression model shows that information asymmetry or excessive transaction friction has a large negative effect on utilisation, then we need to optimise the operating mechanism by establishing simple verification procedures and standardised swap rules to reduce structural obstacles. Through systematic mapping of operational adjustments to the empirically verified regression coefficients, the swap station will give priority to interventions that can maximise the return on resource investment. Due to the strict alignment, the administration's work and limited operational funds will not be scattered among the statistically insignificant variables; instead, they will focus on changing the core mechanisms that affect the behaviour of users and the circulation of resources in practice. Finally, such targeted operation refinement will be used to turn the statistical results of the operation into practical management measures and significantly improve the system's efficiency and sustainability of the campus secondhand book swap station.

According to the regression analysis results, the negative correlation between the distance from the dormitory area and the reuse rate shows that there is a close relationship between them. According to the basic principles, it is necessary to reduce the transaction cost in kind and time of donation and exchange; thus, the threshold of behaviour for students will be lowered. According to the first principle, some facilities need to be improved. Based on the regression results of the spatial layout optimization, it can be seen that beyond a certain distance, the number of participants decreases rapidly. As a result, it is necessary to build 2 to 3 small self-service book-taking points within the central cluster of dormitories that are relatively remote from the main swap station. The scattered satellite nodes can reduce the spatial friction shown in the model and offer easy access points that meet students' daily travel needs.

According to the regression results of the quality of the facility and its ease of operation, new rules should be set for the construction of physical infrastructure. To improve the willingness to participate, a standardised and user-friendly self-service sorting and cataloguing machine is required in the main exchange station and satellite points. It is convenient for students to learn more about participation. The smart deposit lockers at the satellite facility can resist strong winds and rain and be opened at all times; therefore, there is no limitation on the operating time.

To conduct these auxiliary facility improvement rules well, spread out the spatial layout, reduce the distance in kind, and improve supporting facilities to reduce the friction in interaction; then, we can fundamentally change the campus resource circulation ecology of the swap station. Based on the empirical regression results, after these targeted adjustments, the participation threshold is lowered in structure and can contribute to the overall improvement of the second-hand book resource level.

The long-term resource scheduling plan is essentially a dynamic, cyclical system that aims to extend the life cycle of secondhand books and maximize their use. The first principle of this system is based on the regression-validated finding that a high volume of static inventory and time mismatch will lead to a low utilisation rate of resources. Thus, operating rules should not only consider localised short-term storage but also be extended to cover various aspects over a long period. According to the regression coefficients of all significant influencing factors, the plan will be implemented in three interrelated ways. First, to solve the problem of localized idle inventory, a cross-campus second-hand book mutual transfer system has been set up. If a book has been left at a single swap station for more than one semester without any trade, the scheduling algorithm will automatically trigger an inter-station redistribution for that book. It will provide stable supply to all areas of the school with different demand levels and help improve the efficiency of the use of idle stock in some areas. Second, based on the large degree of time variation in academic demand shown by the regression model, a staged book renewal and destocking plan has been made that is strictly aligned with the academic year. At the beginning of the semester, the scheduling priority is given to high-demand course references to promote rapid turnover. As the semester proceeds to mid-term and final, the operational focus shifts to the destocking of old materials through targeted promotional activities to avoid end-of-semester inventory bottlenecks and ensure operational smoothness. Third, to keep the scheduling plan flexible over the long term, a regular user demand survey mechanism has been set up. Based on the updated preferences and reading trends, the feedback loop will adjust the parameters of the regression model and modify the scheduling algorithm in real-time. This repeated optimisation will help the scheduling plan adapt to the changing needs of students. In actual operation, this is a comprehensive long-term arrangement; it will change the swap station from a place of storage to a new node in the flow of resources. Through systematically eliminating the temporal and spatial obstacles to exchange, the plan can gradually enhance the efficiency of resource use at the swap station over time and promote the sustainable circulation of academic resources.

Regression analysis of campus second-hand book swap stations has yielded valuable empirical data on the factors influencing the efficiency of resource use ^[5]. At the basic level, the degree of use of deposited academic resources in this situation is the ratio of how many of them are reclaimed and re-issued by students; this will reflect the work of the book drop-off centre. The main idea of this kind of utilisation mechanism is that in the local education environment, different types of supply and demand will be distributed evenly, theoretically reducing information asymmetry and transaction costs of traditional bilateral book exchanges.

Through the construction and verification of multiple linear regression models in a strict manner, the operation mechanism and development path of the campus second-hand book swap station have been revealed in detail. Based on the regression results, it can be concluded that the degree of newness of the repository, the orderliness of the catalogue, and the convenience of use all have a positive influence on the utilisation rate. On the other hand, the accumulation of inventory is stagnant, and the system will decline in efficiency with time when the unexchanged materials are not used for a long time. According to the empirical results, it is evident that we need to apply this in reality. Campus management can improve circulation bottlenecks by adjusting retrieval algorithms and placing them in different places according to the number of students at various times of the day.

Therefore, the ability to predict in the regression model can be utilised to allocate resources in advance. Based on the relationship between the model and the real data, policies will adjust acquisition standards and automatic weeding rules to maintain a balance between the number of new deposits and old withdrawals. At last, through the systematic application of regression analysis in the management model of the second-hand book swap station, a change has been made in resource circulation from randomness to a predictable, organised form; thus, it has improved the sustainability of the campus circular economy and maximised the use of the left-behind academic resources on campus.

[1]王文婷, 张春月, 刘兵. 高校二手书屋资源管理问题及策略探讨[J]. 现代管理, 2025, 15(2): 75-80.

[2]魏晓梅, 杨运杰. 基于移动端的高校换书平台的开发与应用[J]. 数字技术与应用, 2026, 44(2): 204-206.

[3]董扬媚, 张晨, 曹智. 基于用户画像的二手书交易平台设计与实现[J]. 消费电子, 2025(12): 248-250.

[4]Smith, J., Brown, A.. Advances in Campus Secondhand Book S... (1)[J]. Journal of Interdisciplinary Studies, 2018, 18(4): 47-87.

[5]Smith, J., Brown, A.. Advances in Campus Secondhand Book S... (2)[J]. Journal of Interdisciplinary Studies, 2022, 28(1): 70-84.

[6]Smith, J., Brown, A.. Advances in Campus Secondhand Book S... (3)[J]. Journal of Interdisciplinary Studies, 2019, 17(4): 40-84.