An intelligent express cabinet site selection method based on IF-WSP

Abstract

This invention discloses a smart parcel locker location selection method based on IF-WSP. It utilizes interval fuzzy numbers to establish a fuzzy evaluation matrix, defines random floating parameters, and uses these parameters to weight and integrate expert opinions. An improved chaotic mapping function based on nonlinear perturbation is employed. The method calculates cumulative and summative location selection schemes based on evaluation standard weights, generates a chaotic threshold using the nonlinear perturbation chaotic mapping function, and uses this threshold to perform multiple synthesis of the cumulative and summative location selection schemes. The average location score is then used for defuzzification to determine the final location result. Compared with existing technologies, this invention uses fuzzy logic to transform expert opinions into mathematical expressions and combines them with weighted aggregation and multiplication, fully considering the satisfaction of different customers. This results in parcel locker locations that are closer to customer needs, improving the performance of the parcel lockers and demonstrating significant social value and market application prospects.

Description

Technical Field

[0001] This invention relates to the field of parcel locker site selection, specifically to a smart parcel locker site selection method based on IF-WSP. Background Technology

[0002] With the improvement of residents' consumption level and the booming development of e-commerce in my country, the demand for express delivery has been increasing year by year. Although the scale of the express delivery industry continues to expand, the blind investment in express delivery lockers has led to problems such as high costs and poor user experience. At the same time, the construction of the last-mile delivery system has not kept pace, forming an increasingly prominent contradiction. As a result, the "last mile" delivery problem is quite prominent, such as insufficient express delivery personnel, damaged packages, and inadequate delivery services. Among these problems hindering the development of the express delivery industry, facility site selection is crucial to logistics. This issue has profoundly affected the operation of the express delivery industry. Moreover, with the acceleration of urbanization and the continuous increase in population, urban areas have experienced delivery congestion and an increase in customer complaints. This places higher demands on express delivery companies in the logistics site selection process, requiring more reasonable theories to support the development of last-mile logistics, thereby improving the quality and service effect of logistics delivery and reducing logistics delivery costs.

[0003] The location of smart parcel lockers largely determines their cost. Once the location and scale are determined, significant investment is required to maintain operations, with a long investment cycle and slow cost recovery. Subsequent renovations or demolitions also incur substantial costs. Currently, many researchers, both domestic and international, have conducted in-depth studies on the location selection problem in express delivery networks, particularly focusing on the site layout and route optimization of last-mile logistics delivery. However, common research mainly concentrates on ensemble coverage models and largely relies on subjective judgment or known data to determine parcel locker locations. These studies typically focus on minimizing costs or maximizing customer satisfaction, which has certain limitations. Summary of the Invention

[0004] Purpose of the invention: To address the problems in the background technology, this invention discloses an intelligent express locker location selection method based on IF-WSP. It combines fuzzy logic with weighted summation and accumulation, and generates a chaotic threshold by a nonlinear perturbation chaotic mapping function. This ensures that the method focuses more on the service level of the express delivery industry itself, fully considers the satisfaction of different customers, and better describes the dynamic characteristics of the entire system. This makes the selected express locker location closer to customer needs, improves the performance of the express locker, and has significant social value and market application prospects.

[0005] Technical solution: This invention discloses a method for selecting the location of intelligent express lockers based on IF-WSP, including the following steps:

[0006] Step 1: Obtain multiple smart parcel locker areas for site selection and evaluate them. Establish a fuzzy evaluation matrix using interval fuzzy numbers, initialize the comprehensive number, and define a random floating function. And the expert opinions were weighted and integrated;

[0007] Step 2: Propose an improved chaotic mapping function based on nonlinear perturbation; introduce branches and initial control parameters into the traditional Tent chaotic mapping equation, and introduce a nonlinear function. A chaotic sequence is generated in the interval [0, 1].

[0008] Step 3: Calculate the cumulative location scheme based on the evaluation criteria weights Compared with cumulative site selection schemes The chaotic threshold is generated by the chaotic mapping function in step 2, and the chaotic threshold is used to synthesize the cumulative addressing scheme. and cumulative site selection scheme Update the total number of times;

[0009] Step 4: Average the multiple site selection scores, complete the defuzzing process, calculate and sort the evaluation scores of each expert for different express delivery points, and determine the final site selection result.

[0010] Furthermore, the evaluation criteria for expert opinions in step 1 include B1 sociality, B2 proximity, B3 convenience, and B4 financiality. The evaluation results for the selection of express delivery locker locations are set into 7 levels, namely excellent, good, above average, average, below average, poor, and very poor.

[0011] Furthermore, in step 1, a random floating function is defined. The expert opinions were then weighted and integrated, specifically as follows:

[0012] Random floating function The specific definitions are as follows:

[0013] ;

[0014] in, To describe the problem, The floating parameters describe the problem. , For two membership functions, , and , The upper and lower intervals corresponding to each membership function. The numbers are random, and during each calculation, The value changes over time;

[0015] Through random floating function Calculate floating parameters Complete the weighted integration of expert opinions:

[0016]

[0017] in, For the first Fuzzy evaluation values ​​for each location selection scheme It is the evaluation expert's serial number. It is the evaluation criterion number. The integrated evaluation matrix, For the first The fluctuation parameter of the expert evaluation results Indicates the first The weight of each review expert, It refers to the number of experts.

[0018] Furthermore, the improved chaotic mapping function based on nonlinear perturbation in step 2, on the one hand, introduces branches and initial control parameters into the traditional Tent chaotic mapping equation. , On the other hand, introducing nonlinear functions Through nonlinear functions Control parameters in , The chaotic sequence is generated in the interval [0, 1]. The specific implementation process is as follows:

[0019] ;

[0020] ;

[0021] in, , These are the control parameters, , These are mapping variables In the sequence The value of the nth iteration. For branch control parameters, These are the initial control parameters.

[0022] Furthermore, in step 3, the cumulative location scheme based on the evaluation criterion weights is calculated. Compared with cumulative site selection schemes The details are as follows:

[0023] Step 3.1: Calculate the cumulative location scheme based on the evaluation criteria weights. :

[0024] ;

[0025] in, The total number of evaluation criteria. For the first Each evaluation criterion has weight. The weighted decision matrix of the cumulative location scheme; the weighted decision matrix of the cumulative location scheme. Defined as:

[0026] ;

[0027] in, , For two membership functions, , and , The upper and lower intervals corresponding to each membership function. For the first Each evaluation criterion weight, The integrated evaluation matrix, Number the site selection scheme. The evaluation criterion number;

[0028] Step 3.2: Calculate the cumulative location scheme based on the evaluation criteria weights :

[0029] ;

[0030] in, The total number of evaluation criteria. For the first Each evaluation criterion has weight. The weighted decision matrix for cumulative location selection schemes is defined as follows:

[0031] .

[0032] Furthermore, a chaotic threshold synthesis and summation location scheme is utilized. and cumulative site selection scheme The total number of updates is as follows:

[0033] ;

[0034] in, The chaos threshold, .

[0035] Furthermore, in step 4, the average multiple site selection scores are calculated, defuzzification is performed, the evaluation scores of each expert for different express delivery points are calculated and ranked, and the final site selection result is determined, as follows:

[0036] The deblurring algorithm is as follows:

[0037]

[0038] in, , The membership function is the result of the above calculations. , and , The upper and lower intervals corresponding to each membership function

[0039] Beneficial effects:

[0040] 1. This invention proposes an intelligent parcel locker site selection method based on IF-WSP. This method introduces a decision-making method based on interval fuzzy logic and effectively combines cumulative and summative site selection schemes. By ranking the evaluation scores of various experts for different parcel points, the optimal site selection result is obtained. This further improves the parcel locker entry rate, enhances consumer satisfaction by rationally selecting the location of parcel lockers, thereby expanding the market share of the industry and improving economic and social benefits.

[0041] 2. The nonlinear perturbation chaotic mapping function proposed in this invention adds nonlinear functions. By adjusting the parameters in the function, the nonlinearity and complexity of the system state evolution are increased. In addition, by combining the branch and initial control parameters, the obtained function can have better chaotic characteristics, generate chaotic space in a larger parameter range, and have good security and pseudo-randomness, thus better describing the dynamic characteristics of the entire system.

[0042] 3. This invention, through consultation with various experts in the region, establishes four criteria for the selection of express delivery locker locations: B1 (social aspect), B2 (proximity), B3 (convenience), and B4 (financial aspect). This approach better focuses on the service level of the express delivery industry itself and fully considers the satisfaction of different customers, thereby ensuring that the selected locker locations are closer to customer needs. This guarantees a smoother operation of the express delivery system, significantly improves customer satisfaction, encourages more customers to choose and use smart lockers, and increases the rate of express delivery locker usage.

[0043] 4. This invention designs a random floating function. The expert opinions are weighted and integrated, and the chaos threshold is generated by the chaos mapping function. To integrate the site selection scheme and cumulative site selection scheme Furthermore, by solving the equilibrium location scheme 10 times, the diversity of package locker locations can be improved. That is, by generating random and chaotic parameters within a more refined range to regulate different location schemes, the dynamic characteristics of the entire system can be better described, thereby ensuring more robust and accurate decision analysis, increasing the robustness of package locker location selection decisions, and further improving the applicability and effectiveness of the proposed method. Attached Figure Description

[0044] Figure 1 The implementation process of the intelligent parcel locker site selection method based on IF-WSP;

[0045] Figure 2 This represents the chaotic data distribution generated by the chaotic mapping function based on nonlinear perturbation. Detailed Implementation

[0046] To better explain and facilitate understanding of the present invention, the technical solution of the present invention is described in detail below. The following embodiments are illustrative of the present invention, but the present invention is not limited to the following embodiments.

[0047] This invention proposes a site selection method for intelligent express delivery lockers based on IF-WSP, which utilizes interval fuzzy numbers to establish a fuzzy evaluation matrix and defines a random fluctuation parameter. This paper proposes an improved chaotic mapping function based on nonlinear perturbation to calculate a cumulative location scheme based on evaluation criteria weights. The function is then used to weight and integrate expert opinions. Compared with cumulative site selection schemes A chaotic threshold is generated by a chaotic mapping function, and the chaotic threshold is used to adjust the summation and addressing scheme. and cumulative site selection scheme Ten rounds of aggregation were performed. Fourth, the scores from the ten site selections were averaged, and defuzzification was completed to determine the final site selection result. Finally, an experimental study on the site selection method for smart parcel lockers based on IF-WSP was completed.

[0048] The implementation process of the intelligent parcel locker site selection method based on IF-WSP proposed in this invention is as follows: Figure 1 As shown, the specific steps are as follows:

[0049] In the implementation process, a study area near a residential community was selected as the research area for the parcel locker location problem. A total of 7 locations (S1-S7) were selected, and the proposed IF-WSP-based parcel locker location method was used to evaluate the locations of each parcel locker. To verify the effectiveness of the proposed method, an expert group of 5 experts (Experts 1-5) was established. The selection criteria for experts included two aspects: first, experts with master's or doctoral degrees specializing in parcel locker facility location selection, familiar with the requirements of various aspects of parcel locker location selection, and having a holistic consideration of all influencing factors; second, long-term users who frequently use parcel lockers, whose long-term usage experience has a guiding role in the selection of parcel locker locations. The experts needed to provide evaluation criteria for each parcel locker location, determine the weights, and specifically evaluate the 7 parcel locker locations. Then, according to the above method, the evaluation ranking of the 7 locations and the corresponding cumulative scores were obtained. That is, each expert would receive an evaluation score for the 7 locations. Finally, the top four locations with the highest cumulative score or average score from the expert evaluation results were selected as the optimal locations for parcel locker selection.

[0050] The following example illustrates the evaluation process for one of the delivery locations:

[0051] Step (1): Establish a fuzzy evaluation matrix using interval fuzzy numbers, initialize the comprehensive number, and define the random floating parameter. And it uses this to weight and integrate expert opinions, as follows:

[0052] Step 1.1: Establish a fuzzy evaluation matrix using interval fuzzy numbers.

[0053] Four evaluation criteria and their weights were determined: B1 (sociality), B2 (proximity), B3 (convenience), and B4 (financiality). See Table 1:

[0054] Table 1 Evaluation Criterion Weights

[0055]

[0056] Then, based on actual expert opinions, the evaluation results for the selection of express delivery locker locations were set at 7 levels: Excellent, Good, Above Average, Average, Below Average, Poor, and Very Poor, as shown in Table 2. , For two membership functions, , and , The upper and lower intervals are defined for each membership function. Experts define the location of each parcel locker according to the terminology given in Table 2, establish the corresponding fuzzy evaluation matrix using interval fuzzy numbers, and construct candidate evaluation matrices for each expert to compare the merits of the selected parcel locker locations. The overall iteration count is initialized to 10.

[0057] Table 2 Membership Degrees Represented by Interval Fuzzy Numbers

[0058]

[0059] This involves weighting and integrating expert opinions, using... That is, the first The weighting of each review expert is determined in practice based on their familiarity with the express delivery industry. Specifically, Specialists 1-3 are master's and doctoral degree experts specializing in the site selection of parcel lockers. These experts have a holistic view of all influencing factors in parcel locker site selection, therefore, their weight is slightly higher. Specialists 4-5 are user experts who frequently use parcel lockers. Their long-term experience can guide the site selection of parcel lockers, but they are not familiar with some of the more technical details of site selection, therefore, their weight is set slightly lower. The specific settings are as follows:

[0060] Table 3. Weighting of Review Experts

[0061]

[0062] Step 1.2: Define the random floating function

[0063] The specific definition of a floating function is as follows:

[0064] (1)

[0065] in, To describe the problem, The floating parameters describe the problem. , For two membership functions, , and , The upper and lower intervals corresponding to each membership function. The numbers are random, and during each calculation, The values ​​vary between intervals.

[0066] Step 1.3: Perform weighted integration of expert opinions based on floating parameters.

[0067] Through random floating function Calculate floating parameters The following algorithm is used to weight and integrate expert opinions:

[0068] (2)

[0069] in, For the first Fuzzy evaluation values ​​for each location selection scheme It is the evaluation expert's serial number. It is the evaluation criterion number. The integrated evaluation matrix, For the first The fluctuation parameter of the expert evaluation results Indicates the first The weight of each review expert, It refers to the number of experts.

[0070] Step (2) implements an improved chaotic mapping function based on nonlinear perturbation.

[0071] An improved chaotic mapping based on nonlinear perturbation is designed. The idea is to introduce branches and initial control parameters into the traditional Tent chaotic mapping equation. , On the other hand, nonlinear functions were introduced. Through the adjustment parameters in this function , This generates a chaotic sequence in the interval [0, 1], thereby increasing the diversity of solutions. The specific implementation process is as follows:

[0072] (3)

[0073] (4)

[0074] in, , These are the control parameters, , These are mapping variables In the sequence The value of the nth iteration. For branch control parameters, For the initial control parameters, it should be noted that when When the value is 0.5, it should be avoided. The system exhibits a short-period state when the value falls within (0.2, 0.4, 0.6, 0.8), or when it falls within (0, 0.25, 0.5, 0.75), resulting in a fixed point. Experiments have verified that setting the parameters... for .

[0075] The chaotic data distribution generated by the chaotic mapping function based on nonlinear perturbation is as follows: Figure 2 .

[0076] Step (3) Calculate the cumulative location scheme based on the evaluation criteria weights Compared with cumulative site selection schemes ; Using chaotic threshold summation location scheme and cumulative site selection scheme Update the total number of iterations until 10 iterations are completed, as detailed below:

[0077] Step 3.1: Calculate the cumulative location scheme based on the evaluation criteria weights.

[0078] The formula for calculating the cumulative location scheme is as follows:

[0079] (5)

[0080] in, The total number of evaluation criteria. For the first Each evaluation criterion has weight. It is the weighted decision matrix of the summation and location schemes.

[0081] Weighted decision matrix of cumulative location schemes Defined as:

[0082] (6)

[0083] in, For the first Each evaluation criterion weight, The integrated evaluation matrix, Number the site selection scheme. This is the evaluation criterion number.

[0084] In the calculation process, taking the first and second evaluation criteria as examples, the calculation formula for the cumulative term is shown below:

[0085] (7)

[0086] In the cumulative location scheme, the upper and lower intervals corresponding to the membership function used by each expert are... , and , As shown in Table 4:

[0087] Table 4. Additive Site Selection Scheme of Five Experts membership degree

[0088]

[0089] Step 3.2: Calculate the cumulative location scheme based on the evaluation criteria weights

[0090] The cumulative location scheme calculation formula is as follows:

[0091] (8)

[0092] in, The total number of evaluation criteria. For the first Each evaluation criterion has weight. This is the weighted decision matrix for the cumulative location selection schemes.

[0093] The weighted decision matrix of the cumulative location scheme is defined as follows:

[0094] (9)

[0095] Taking the first and second evaluation criteria as examples, the calculation formula for their cumulative terms is shown below:

[0096] (10)

[0097] In the multiplicative addressing scheme, the upper and lower intervals corresponding to the membership functions used by each expert are... , and , As shown in Table 5:

[0098] Table 5. Multiplicative Site Selection Schemes of Five Experts membership degree

[0099]

[0100] Step 3.3: Utilizing the Chaos Threshold Accumulation and site selection scheme and cumulative site selection scheme Complete 10 comprehensive processing steps. Details are as follows:

[0101] (11)

[0102] in, The chaos threshold, .

[0103] During implementation, the above formula is used to synthesize and sum the site selection scheme. and cumulative site selection scheme The resulting combined model.

[0104] Step (4): Calculate the average score of ten site selections, perform defuzzing, calculate and rank the evaluation scores of each expert for different express delivery points, and determine the final site selection results, as follows:

[0105] The average score from ten site selection trials is used to perform deblurring. Then, the highest-scoring scheme is determined as the final site selection result. The deblurring algorithm is as follows:

[0106] (12)

[0107] in, , The membership function is the result of the above calculations. , and , The upper and lower intervals corresponding to each membership function.

[0108] Step (5) completes the experimental study on the intelligent express cabinet location selection method based on IF-WSP.

[0109] Following the implementation process of the above embodiments, each expert evaluated each of the 7 express delivery points one by one, and determined their final cumulative score and average score for all express delivery locker locations. The final ranking of the solutions is shown in Table 6.

[0110] Table 6. Ranking of the final solutions from the five experts.

[0111] expert The order of scores for the location of the parcel locker (from highest to lowest) Average score Sort Specialized 3 S1- S5- S2- S7- S3- S6- S4 0.89 1 Specialty 2 S1- S5- S2- S7- S3- S6- S4 0.71 2 Specialty 5 S1- S5- S2- S7- S3- S6- S4 0.65 3 Specialist 1 S1- S5- S2- S7- S3- S6- S4 0.6 4 Specialized 4 S1- S5- S2- S7- S3- S6- S4 0.54 5

[0112] As shown in Table 6, Special Project 3 provides the best ranking of parcel locker locations, with a final average score of 0.89. Special Project 4 provides the worst ranking of parcel locker locations, with an average score of 0.54. The rankings of parcel locker locations selected by Special Projects 2, 5, and 1 are in the middle. Therefore, the ranking of the 7 locations selected by Special Project 3 (S1-S5-S2-S7-S3-S6-S4) is the most reliable, and the first 4 locations (S1-S5-S2-S7) are determined to be the best site selection scheme for the parcel lockers.

[0113] Finally, for schemes 1-5, 2000 residents were randomly selected from the study area to complete a questionnaire survey. Each resident was asked to select 4 locations from 7 express delivery point locations as the best express delivery locker locations. The specific results are shown in the table below:

[0114] Table 7 Questionnaire Survey Results

[0115]

[0116] As shown in Table 7, the top four suggestions that support S1, S5, S2, and S7 as the optimal locations are supported by 1540, 1393, 1412, and 1123 people respectively. This is consistent with the conclusion obtained from the proposed method for selecting the location of the parcel locker. Therefore, this method for selecting the location of the parcel locker meets the user's requirements.

[0117] The above embodiments are only for illustrating the technical concept and features of the present invention, and are intended to enable those skilled in the art to understand the content of the present invention and implement it accordingly. They should not be construed as limiting the scope of protection of the present invention. All equivalent transformations or modifications made in accordance with the spirit and essence of the present invention should be covered within the scope of protection of the present invention.

Claims

1. A method for site selection of intelligent parcel lockers based on IF-WSP, characterized in that, Includes the following steps: Step 1: Obtain multiple smart parcel locker areas for site selection and evaluate them. Establish a fuzzy evaluation matrix using interval fuzzy numbers, initialize the comprehensive number, and define a random floating function. And the expert opinions were weighted and integrated; Step 2: Propose an improved chaotic mapping function based on nonlinear perturbation; introduce branches and initial control parameters into the traditional Tent chaotic mapping equation, and introduce a nonlinear function. A chaotic sequence is generated in the interval [0, 1]. Step 3: Calculate the cumulative location scheme based on the evaluation criteria weights Compared with cumulative site selection schemes The chaotic threshold is generated by the chaotic mapping function in step 2, and the chaotic threshold is used to synthesize the cumulative addressing scheme. and cumulative site selection scheme Update the total number of times; Step 4: Average the multiple site selection scores, complete the defuzzing process, calculate and sort the evaluation scores of each expert for different express delivery points, and determine the final site selection result.

2. The intelligent express locker site selection method based on IF-WSP according to claim 1, characterized in that, The evaluation criteria for expert opinions in step 1 include B1 sociality, B2 proximity, B3 convenience, and B4 financiality. The evaluation results of the parcel locker site selection are set into 7 levels, namely excellent, good, above average, average, below average, poor, and very poor.

3. The intelligent parcel locker site selection method based on IF-WSP according to claim 1, characterized in that, In step 1, a random floating function is defined. The expert opinions were then weighted and integrated, specifically as follows: Random floating function The specific definitions are as follows: ； in, To describe the problem, The floating parameters describe the problem. , For two membership functions, , and , The upper and lower intervals corresponding to each membership function. The number is random, and during each calculation, The value changes over time; Through random floating function Calculate floating parameters Complete the weighted integration of expert opinions: ； in, For the first Fuzzy evaluation values ​​for each location selection scheme It is the evaluation expert's serial number. It is the evaluation criterion number. The integrated evaluation matrix, For the first The fluctuation parameter of the expert evaluation results Indicates the first The weight of each review expert, It refers to the number of experts.

4. The intelligent express locker site selection method based on IF-WSP according to claim 1, characterized in that, The improved chaotic mapping function based on nonlinear perturbation in step 2 introduces branches and initial control parameters into the traditional Tent chaotic mapping equation. , On the other hand, introducing nonlinear functions Through nonlinear functions Control parameters in , The chaotic sequence is generated in the interval [0, 1]. The specific implementation process is as follows: ； ； in, , These are the control parameters, , These are mapping variables In the sequence The value of the nth iteration. For branch control parameters, These are the initial control parameters.

5. The intelligent express locker site selection method based on IF-WSP according to claim 1, characterized in that, In step 3, the cumulative location scheme based on the evaluation criteria weights is calculated. Compared with cumulative site selection schemes The details are as follows: Step 3.1: Calculate the cumulative location scheme based on the evaluation criteria weights. : ； in, The total number of evaluation criteria. For the first Each evaluation criterion has weight. The weighted decision matrix of the cumulative location scheme; the weighted decision matrix of the cumulative location scheme. Defined as: ； in, , For two membership functions, , and , The upper and lower intervals corresponding to each membership function. For the first Each evaluation criterion weight, The integrated evaluation matrix, Number the site selection scheme. The evaluation criterion number; Step 3.2: Calculate the cumulative location scheme based on the evaluation criteria weights : ； in, The total number of evaluation criteria. For the first Each evaluation criterion has weight. The weighted decision matrix for cumulative location selection schemes is defined as follows: 。 6. The intelligent express locker site selection method based on IF-WSP according to claim 4, characterized in that, Chaotic threshold synthesis summation location scheme and cumulative site selection scheme The total number of updates is as follows: ； in, The chaos threshold, .

7. The intelligent express locker site selection method based on IF-WSP according to claim 1, characterized in that, In step 4, the average multiple site selection scores are calculated, defuzzing is performed, the evaluation scores of each expert for different express delivery points are calculated and ranked, and the final site selection result is determined, as follows: The deblurring algorithm is as follows: ； in, , The membership function is the result of the above calculations. , and , The upper and lower intervals corresponding to each membership function.

Images