An order dispatching method and device, electronic equipment and readable storage medium

Abstract

Embodiments of the present application provide an order distribution method and device, electronic equipment and readable storage medium, the method comprises: acquiring marketing data, acquiring at least one index for optimizing target order from marketing data;Acquire the weight of each index, and at least one index data of at least one pre-distribution object for at least one index;According to at least one index data of at least one pre-distribution object and the weight of each index, generate the rank sum ratio of each pre-distribution object;Rank sum ratio is used to indicate the evaluation degree of pre-distribution object;According to the rank sum ratio of each pre-distribution object, determine target order receiving object.Through the rank sum ratio algorithm, the plurality of personalized factors of the operator product marketing are considered comprehensively, and based on the comprehensive evaluation ranking result indicated by the rank sum ratio for the plurality of pre-distribution objects, a certain order is distributed to the best order receiving object, so that each distributed order can get the best converted business demand.

Description

Technical Field

[0001] This invention relates to the field of communication technology, and in particular to an order dispatch method, an order dispatch device, a corresponding electronic device, and a corresponding computer-readable storage medium. Background Technology

[0002] In the telecommunications industry, due to the unique nature of communication services, most online orders require dispatching to offline channels for marketing processing. Offline channel marketing processing involves the recipients of the orders, and the dispatching requirements typically consider order conversion rates, which are influenced by relevant metrics.

[0003] In the related technologies of order assignment, the main methods are to determine the order recipient based on a single indicator comparison by the order dispatcher (i.e., the order distributor), or to determine the order recipient by manually determining the indicator weight and then calculating the score. However, the order recipients determined by the aforementioned methods are greatly affected by the subjective consciousness of the order dispatcher, which may result in a large error in the evaluation results and cannot guarantee the selection of the best order recipient. Summary of the Invention

[0004] In view of the above problems, embodiments of the present invention are proposed to provide an order dispatching method, an order dispatching device, a corresponding electronic device, and a corresponding computer-readable storage medium to overcome or at least partially solve the above problems.

[0005] This invention discloses an order dispatch method, the method comprising:

[0006] Acquire marketing data, and extract at least one metric from the marketing data for optimizing target orders; wherein the marketing data includes marketing process data and marketing result data;

[0007] Obtain the weights of each indicator, and at least one indicator data for at least one pre-assigned object for the at least one indicator;

[0008] Based on at least one indicator data of the at least one pre-assigned object and the weight of each indicator, the rank-sum ratio of each pre-assigned object is generated; the rank-sum ratio is used to indicate the degree of excellence of the pre-assigned object.

[0009] The target order recipient is determined based on the rank-sum ratio of each pre-assigned object.

[0010] Optionally, obtaining the weights of each indicator includes:

[0011] If the indicator is a preset first type of indicator, then the weight of the preset first type of indicator is determined by the preset hierarchical analysis method.

[0012] And / or, if the indicator is a preset second type of indicator, then the weight of the preset second type of indicator is determined by the preset entropy weight method.

[0013] Optionally, determining the weights of the preset first type of index using a preset analytic hierarchy process includes:

[0014] Obtain at least one of the preset first-category indicators;

[0015] The weights of each preset first-category indicator are determined based on the comparison results of at least one pairwise indicator among the preset first-category indicators.

[0016] Optionally, determining the weights of the preset second type of index using the preset entropy weight method includes:

[0017] Obtain the information entropy of each preset second-type indicator, and use the information entropy to calculate the weight of each preset second-type indicator.

[0018] Optionally, generating the rank-sum ratio of each pre-assignment object based on at least one indicator data of the at least one pre-assignment object and the weights of each indicator includes:

[0019] Based on at least one indicator data of the at least one pre-assigned object, the rank of each indicator of each pre-assigned object is obtained;

[0020] The rank-sum ratio of each pre-assigned object is generated based on the rank of each indicator and the weight of each indicator.

[0021] Optionally, obtaining the rank of each indicator of each pre-assignment object based on at least one indicator data of the at least one pre-assignment object and the weight of each indicator includes:

[0022] A data matrix is ​​generated using the number of metrics used to optimize the target order and the number of pre-assigned objects; wherein each column of the data matrix indicates each metric used to optimize the target order, and each row of the data matrix indicates each pre-assigned object;

[0023] Based on the data of each indicator for each pre-assigned object, the rank of each indicator of each pre-assigned object in the data matrix is ​​ranked to obtain a rank matrix; wherein, the rank matrix contains the rank of each indicator of each pre-assigned object.

[0024] Optionally, the rank-sum ratio is positively correlated with the evaluation level of the pre-assigned objects; the step of determining the target order recipient based on the rank-sum ratio of each pre-assigned object includes:

[0025] Sort each pre-assignment object according to the rank-sum ratio of each pre-assignment object;

[0026] Based on the sorting results, the target order recipients are determined.

[0027] This invention also discloses an order dispatching device, the device comprising:

[0028] The indicator acquisition module is used to acquire marketing data and extract at least one indicator from the marketing data for optimizing target orders; wherein the marketing data includes marketing process data and marketing result data;

[0029] The indicator data acquisition module is used to acquire the weight of each indicator and at least one indicator data of at least one pre-assigned object for the at least one indicator.

[0030] The rank-sum ratio generation module is used to generate the rank-sum ratio of each pre-assignment object based on at least one indicator data of the at least one pre-assignment object and the weight of each indicator; the rank-sum ratio is used to indicate the degree of excellence of the pre-assignment object.

[0031] The order receiving object determination module is used to determine the target order receiving object based on the rank-sum ratio of each pre-assigned object.

[0032] Optionally, the indicator data acquisition module includes:

[0033] The weight acquisition submodule is used to determine the weight of the preset first type of indicator by using a preset analytic hierarchy process when the indicator is a preset first type of indicator; and / or, to determine the weight of the preset second type of indicator by using a preset entropy weight method when the indicator is a preset second type of indicator.

[0034] Optionally, the weight acquisition submodule includes:

[0035] The weight acquisition unit is used to acquire at least one of the preset first-class indicators, determine the weight of each preset first-class indicator based on the comparison results of pairs of indicators among the at least one preset first-class indicators, and / or acquire the information entropy of each preset second-class indicator, and calculate the weight of each preset second-class indicator using the information entropy.

[0036] Optionally, the rank-to-ratio generation module includes:

[0037] The rank-sum ratio generation submodule is used to obtain the rank of each indicator of each pre-assignment object based on at least one indicator data of the at least one pre-assignment object; and to generate the rank-sum ratio of each pre-assignment object based on the rank of each indicator of each pre-assignment object.

[0038] Optionally, the rank-to-ratio generation submodule includes:

[0039] A rank generation unit is used to generate a data matrix using the number of metrics for optimizing the target order and the number of pre-assignment objects; wherein each column of the data matrix indicates each metric for optimizing the target order, and each row of the data matrix indicates each pre-assignment object; based on the metric data of each pre-assignment object for each metric, each metric of each pre-assignment object in the data matrix is ​​ranked to obtain a rank matrix; wherein the rank matrix contains the rank of each metric of each pre-assignment object.

[0040] Optionally, the rank-sum ratio is positively correlated with the evaluation level of the pre-assigned object; the order-receiving object determination module includes:

[0041] The order receiving object determination submodule is used to sort each pre-assigned object according to the rank-sum ratio of each pre-assigned object; and determine the target order receiving object based on the sorting result.

[0042] This invention also discloses an electronic device, including: a processor, a memory, and a computer program stored in the memory and capable of running on the processor, wherein the computer program, when executed by the processor, implements any of the order dispatch methods described above.

[0043] This invention also discloses a computer-readable storage medium storing a computer program, which, when executed by a processor, implements any of the order dispatch methods described above.

[0044] The embodiments of the present invention have the following advantages:

[0045] In this embodiment of the invention, at least one indicator for optimizing target orders is obtained from marketing process data and marketing result data, and the weight of each indicator is obtained. Then, at least one pre-assignment object is used. Based on the at least one indicator data and the weight of each indicator, the rank-sum ratio of each pre-assignment object is obtained. This rank-sum ratio can be mainly used to indicate the degree of excellence of the pre-assignment object. At this time, the target order receiving object to be assigned to the target order can be determined based on the rank-sum ratio of each pre-assignment object. By comprehensively considering multiple personalized factors of operator product marketing through the rank-sum ratio algorithm, and based on the ranking result of multiple pre-assignment objects indicated by the rank-sum ratio, it is used to indicate the individual comprehensive evaluation ranking of each pre-assignment object, and to assign a certain order to the best receiving object, so that each assigned order can achieve the best conversion business needs. Attached Figure Description

[0046] Figure 1 This is a flowchart illustrating the steps of an embodiment of the order dispatch method of the present invention;

[0047] Figure 2This is a flowchart illustrating the steps of another embodiment of the order dispatch method of the present invention;

[0048] Figure 3 This is a schematic diagram illustrating an application scenario of order dispatch provided in an embodiment of the present invention;

[0049] Figure 4 This is a structural block diagram of an embodiment of an order dispatching device according to the present invention. Detailed Implementation

[0050] To make the above-mentioned objects, features and advantages of the present invention more apparent and understandable, the present invention will be further described in detail below with reference to the accompanying drawings and specific embodiments.

[0051] In the telecommunications industry, offline channel marketing involves order recipients, and order dispatch typically considers order conversion rates. Order conversion is influenced by several metrics, such as whether the order was accepted for marketing purposes; higher order prices lead to greater short-term profits; for bundled packages (combining mobile numbers with other products like broadband, iTV, and smart home services), better customer stability translates to greater long-term profits; and better service enhances brand image. The aforementioned order conversion rates are dependent on the capabilities and geographical proximity of the offline order recipients.

[0052] In order dispatching technologies, one example is the use of a single metric by the dispatcher (order distributor) to determine order recipients. However, this method is heavily influenced by the dispatcher's subjective biases, lacking a scientific and objective evaluation. Consequently, conversion rates vary widely, failing to guarantee overall order optimization. For instance, if the dispatcher selects recipients solely based on distance from the customer, a poor sales performance by that recipient could compromise order success. Similarly, relying solely on past order success rates might not guarantee customer value (i.e., price). Furthermore, selecting recipients based solely on order price could also compromise order success. A single target may not guarantee customer service satisfaction. Therefore, telecommunications operators usually need to consider multiple indicators. Using only the comparison result of a single indicator is difficult to meet the overall optimal needs and has a significant impact on the company's interests and long-term development. For example, when using multiple indicators for comparison, the weight of the indicators can be determined by humans to calculate the score and decide the order recipient. However, the above method mainly relies on subjective judgment to determine the weight of all indicators. It depends on the business expertise of the person determining the indicators, which may lead to large errors in the evaluation results. Moreover, it may not be possible to determine the corresponding weight of indicators for which there is no prior business experience, and it cannot guarantee the selection of the best order recipient.

[0053] This invention comprehensively considers multiple personalized factors in operator product marketing through the rank-sum ratio algorithm, thereby achieving a multi-dimensional comprehensive evaluation of pre-assignment objects. Based on the ranking results of multiple pre-assignment objects indicated by the rank-sum ratio, it guides the individual comprehensive evaluation ranking of each pre-assignment object, assigning an order to the best receiving object. This ensures that each assigned order meets the optimal business needs for conversion, thereby improving the overall conversion efficiency of orders. Furthermore, the calculation of indicator weights can be conducted scientifically, combining qualitative and quantitative methods. For example, the qualitative part can use the analytic hierarchy process (AHP) to reduce errors caused by subjective judgment, while the quantitative part can use the entropy weight method to evaluate weights from the perspective of data information, making the indicators more meaningful. This allows for quantitative analysis of qualitative issues, improving the scientific rigor of the evaluation results and reducing errors caused by human subjectivity compared to manually assigning weights. Finally, by using the rank-sum ratio comprehensive evaluation model, the most suitable receiving object is selected, ensuring that the decision factor for order assignment focuses on the ranking of the pre-assignment object rather than the score itself. In addition, the selection of evaluation indicators can be based on three types of data: the order results data of the order recipient, the order marketing process data, and its own data. Evaluation indicators can be mined from the marketing process data and marketing results data and reflected in the comprehensive evaluation results, making the marketing process of the selected order recipients more standardized and regulated.

[0054] Reference Figure 1 The diagram illustrates a flowchart of an embodiment of the order dispatch method of the present invention, which may specifically include the following steps:

[0055] Step 101: Obtain marketing data and extract at least one metric from the marketing data to optimize target orders;

[0056] An order can refer to any order that needs to be allocated in different fields or application scenarios. In this embodiment of the invention, an order can refer to a business opportunity order, which is information allocated online by a telecommunications operator but not yet processed offline.

[0057] In this embodiment of the invention, the actual business problem of order dispatch can be transformed into a mathematical problem. By establishing a suitable algorithm or model, a predetermined goal can be evaluated or predicted. Specifically, this problem can be a comprehensive evaluation problem with multiple indicators. In this embodiment of the invention, the rank-sum ratio algorithm can be used for comprehensive evaluation.

[0058] The Rank-sum ratio (RSR) is a method that performs a comprehensive evaluation by ranking each indicator based on different objects.

[0059] In one embodiment of the present invention, in order to achieve a comprehensive evaluation of multiple indicators, marketing data and target orders to be assigned can be obtained first. At this time, at least one indicator for optimizing target orders can also be obtained from the marketing data, so as to ensure the optimization of the overall conversion efficiency of target orders based on the comprehensive evaluation of the at least one indicator.

[0060] It should be noted that in this embodiment of the invention, the actual business problem of order dispatch is transformed into a mathematical problem, which can be implemented using a comprehensive evaluation model designed according to subsequent process steps. For the model, the indicator data determines the upper limit of the model. This embodiment of the invention focuses on adjusting some model parameters in a publicly available and transparent model algorithm. The adjustment of some model parameters depends on indicator data, which can be obtained by mining marketing data, such as marketing process data and marketing result data.

[0061] In practical applications, the selection of evaluation indicators can be based on past business experience in the operator industry and a large amount of historical data. Key indicators closely related to operator product marketing can be analyzed and mined. For example, evaluation indicators can be selected by combining three types of data: past order result data of the order recipient (i.e., marketing result data of past orders), order marketing process data, and the order recipient's own attribute data. Marketing result data selected based on business expert experience may include, but is not limited to, success rate, integration rate, customer value (i.e., price), smart home rate, and service satisfaction. Marketing elements of the marketing process data can be extracted by analyzing the business process of telecommunications product marketing, based on business specifications and key links. These may include, but are not limited to, second-party agreement completion rate, average order acceptance time, return order rate, first contact time, average return order time, 15-minute order acceptance rate, 3-day return order rate, and order acceptance rate. The order recipient's own attribute data selected based on expert business experience may include, but is not limited to, distance. The relevant data mined can then be used as evaluation indicators. It should be noted that this embodiment of the invention does not limit the specific methods of data analysis and indicator mining.

[0062] Step 102: Obtain the weights of each indicator and at least one indicator data for at least one pre-assigned object for at least one indicator.

[0063] The comprehensive evaluation of multiple indicators mainly refers to the comprehensive evaluation of at least one pre-assigned object. In this case, we can obtain at least one indicator data and the weight of each indicator for at least one pre-assigned object, so as to achieve the comprehensive evaluation of multiple indicators based on at least one indicator data and the weight of each indicator for at least one pre-assigned object.

[0064] In one embodiment of the invention, the comprehensive evaluation of multiple indicators can be indicated by the calculated rank-sum ratio.

[0065] For example, in the telecommunications operator industry, offline channel marketing acceptance involves order recipients. Pre-assigned recipients can refer to all pending order recipients who are capable of marketing acceptance. At least one indicator data refers to the specific indicator data of each pending order recipient corresponding to each evaluation indicator. Weight refers to the relative importance of a certain indicator in the overall evaluation. This embodiment of the invention does not limit this.

[0066] Since multiple evaluation indicators are involved, their weights can be quantified. When calculating the rank-sum ratio, the importance of each indicator can also be considered. In other words, the weights are important parameters in the rank-sum ratio algorithm.

[0067] Weights can clarify the relative importance of each evaluation indicator in the overall evaluation system, determine the relationship between the corresponding evaluation indicators in the indicator system, and ensure the scientific nature of the overall comprehensive evaluation results based on reasonable weight settings. In this embodiment of the invention, the weights of each indicator can be determined by combining the analytic hierarchy process (AHP) and the entropy weight method. For example, the calculation of indicator weights can be done scientifically, combining qualitative and quantitative methods. For instance, the qualitative part can use the AHP to reduce errors caused by subjective judgment, while the quantitative part can use the entropy weight method to evaluate the weights from the perspective of data information, making the indicators more meaningful. This allows for the quantitative analysis of qualitative issues, thereby improving the scientific nature of the evaluation results. Compared to manually assigning weights, this reduces errors caused by human subjectivity. This embodiment of the invention does not limit the specific method for assigning indicator weights.

[0068] Step 103: Generate the rank-sum ratio of each pre-assigned object based on at least one indicator data of at least one pre-assigned object and the weight of each indicator;

[0069] Specifically, the rank-sum ratio of each pre-assigned object can be generated. This rank-sum ratio can be used to indicate the degree of excellence of the pre-assigned object. Thus, the ranking result of multiple pre-assigned objects can be indicated based on the calculated rank-sum ratio. This ranking result can indicate the specific comprehensive evaluation ranking of each pre-assigned object among all pre-assigned objects, thereby realizing a multi-dimensional comprehensive evaluation of the pre-assigned objects.

[0070] In practical implementation, the generation method of the rank-sum ratio can be expressed as follows: First, based on at least one indicator data of at least one pre-assigned object, the rank of each indicator of each pre-assigned object can be obtained. This rank mainly refers to the ranking / ranking of a certain indicator data of a certain pre-assigned object among all pre-assigned objects. Then, based on the rank of each indicator of each pre-assigned object and the weight of each indicator, the rank-sum ratio of each pre-assigned object can be generated. This rank-sum ratio is generated based on the rank of each indicator of each pre-assigned object, which can realize the comprehensive consideration of multiple personalized factors of operator product marketing, and thus realize the multi-dimensional comprehensive evaluation of pre-assigned objects.

[0071] Step 104: Determine the target order recipient based on the rank-sum ratio of each pre-assigned object.

[0072] In one embodiment of the present invention, the rank-sum ratio of each pre-assigned object can be used to indicate the ranking result of the comprehensive evaluation ranking of an individual after considering comprehensive factors. At this time, the target order receiving object can be determined according to the rank-sum ratio of each pre-assigned object.

[0073] Since the rank-sum ratio algorithm can be used to calculate the final comprehensive evaluation ranking of the order receiving objects, that is, the rank-sum ratio can be positively correlated with the evaluation degree of the pre-assigned objects, when selecting the target order receiving object to be assigned to the target order, the pre-assigned object ranked first can be directly selected as the best order receiving object, so that when the best order receiving object handles the marketing of the target order, the order can achieve the best conversion effect.

[0074] In this embodiment of the invention, at least one indicator for optimizing target orders is obtained from marketing process data and marketing result data, and the weight of each indicator is obtained. Then, at least one pre-assignment object is used. Based on the at least one indicator data and the weight of each indicator, the rank-sum ratio of each pre-assignment object is obtained. This rank-sum ratio can be mainly used to indicate the degree of excellence of the pre-assignment object. At this time, the target order receiving object to be assigned to the target order can be determined based on the rank-sum ratio of each pre-assignment object. By comprehensively considering multiple personalized factors of operator product marketing through the rank-sum ratio algorithm, and based on the ranking result of multiple pre-assignment objects indicated by the rank-sum ratio, it is used to indicate the individual comprehensive evaluation ranking of each pre-assignment object, and to assign a certain order to the best receiving object, so that each assigned order can achieve the best conversion business needs.

[0075] Reference Figure 2 The flowchart illustrates another embodiment of the order dispatch method of the present invention, which may specifically include the following steps:

[0076] Step 201: Obtain the rank of each indicator of each pre-assigned object based on at least one indicator data of at least one pre-assigned object;

[0077] In this embodiment of the invention, in order to achieve a comprehensive evaluation of multiple indicators, the target orders to be assigned can first be obtained. At the same time, at least one indicator for optimizing the target orders can also be obtained, so as to ensure the optimization of the overall conversion efficiency of the target orders based on the comprehensive evaluation of the at least one indicator.

[0078] In one embodiment of the present invention, the comprehensive evaluation of multiple indicators is mainly a comprehensive evaluation of at least one pre-assigned object. At this time, at least one indicator data of at least one pre-assigned object for at least one indicator can be obtained, and then the comprehensive evaluation of multiple indicators can be realized based on at least one indicator data of at least one pre-assigned object.

[0079] Specifically, the generation method for the rank-sum ratio can be described as follows: firstly, based on at least one indicator data of at least one pre-assigned object, the rank of each indicator of each pre-assigned object can be obtained. This rank mainly refers to the sorting / ranking of a certain indicator data of a certain pre-assigned object among all pre-assigned objects.

[0080] The Rank-Sum Ratio (RSR) method can include integer rank-sum ratio and non-integer rank-sum ratio. Integer rank-sum ratio directly ranks the corresponding indicator data in ascending or descending order. If the indicator data are the same, an average rank is used, where the ranking is the same for indicators with identical data, and the ranking is based on the average rank of those indicators. Non-integer rank-sum ratio, on the other hand, ranks and scores each indicator by using the distance from its worst-performing data point, and then calculates the final comprehensive score for each indicator using a weighted average of all indicators.

[0081] It should be noted that, in order to improve the shortcomings of the rank-commissioning method of the RSR method, the embodiments of the present invention can use the non-integer rank-sum ratio method to perform subsequent rank-sum ratio calculations, so as to overcome the disadvantage that the RSR method is prone to losing quantitative information of the original index value when performing rank conversion based on the quantitative linear correspondence between the compiled rank and the original index value.

[0082] In practical implementation, a data matrix can be generated using the number of metrics used to optimize the target order and the number of pre-assignment objects. For example, assuming there are n pre-assignment objects, also known as evaluation objects, and m evaluation metrics, an n×m data matrix can be constructed. The columns of the data matrix can indicate the various metrics used to optimize the target order, and the rows of the data matrix can indicate the various pre-assignment objects. That is, it can also be represented as an n-row m-column original data table, achieving standardized processing of the metric data. Then, based on the metric data of each pre-assignment object for each metric, the rank of each metric of each pre-assignment object in the data matrix can be assigned, thereby obtaining a rank matrix based on the rank of each metric of each pre-assignment object.

[0083] Taking the non-integer rank sum ratio method for calculating the rank sum ratio as an example, when ranking each indicator, the ranking method for benefit-type indicators can be calculated using the following formula:

[0084]

[0085] For cost-related indicators, their ranking can be calculated using the following formula:

[0086]

[0087] Among them, R ij It can refer to the rank of the j-th index of the i-th object, X. ij It can refer to the indicator data corresponding to the j-th indicator of the i-th object. The value of n can be a positive integer (i.e., n≥1), and its maximum value can be the number of pre-assigned objects.

[0088] It should be noted that benefit-type indicators can refer to indicators that are positively correlated with the evaluated object, and generally the larger the value, the better; cost-type indicators can refer to indicators that are negatively correlated with the evaluated object, and generally the smaller the value, the better. For specific examples of indicators, the embodiments of this invention do not limit this.

[0089] Step 202: Generate the rank-sum ratio of each pre-assigned object based on the rank of each indicator and the weight of each indicator.

[0090] In one embodiment of the present invention, the rank-sum ratio of each pre-assigned object can be generated based on the rank and weight of each indicator of each pre-assigned object. This rank-sum ratio is generated based on the rank and weight of each indicator of each pre-assigned object, which enables comprehensive consideration of multiple personalized factors in operator product marketing, thereby achieving a multi-dimensional comprehensive evaluation of the pre-assigned object.

[0091] Specifically, since multiple evaluation indicators are involved, the weights of these indicators can be quantified. When calculating the rank-sum ratio, the importance (i.e., weight) of each indicator can be combined with the judgment.

[0092] In practical applications, the calculated rank matrix can represent the scoring matrix of all indicators of the pre-assigned object. For example, assuming there are n pre-assigned objects and m indicators, an n-row m-column scoring list can be obtained. This scoring list can be combined with the weight of each indicator to calculate the overall score of each pre-assigned object, i.e., the rank-to-score ratio, thereby obtaining the ranking result of the individual comprehensive evaluation ranking for each pre-assigned object.

[0093] Specifically, the weights of each indicator can be obtained, and then the rank and weights of each indicator of each pre-assigned object can be used to calculate the rank-sum ratio of each pre-assigned object.

[0094] In practical implementation, the acquisition of indicator weights can be achieved by combining qualitative and quantitative methods to determine the importance of each indicator, that is, to determine the weight of each indicator, so as to quantify the weight of the indicators through numerical proportions.

[0095] As an example, if the indicator is a preset first-class indicator, the preset first-class indicator usually refers to the indicators whose relative importance can be judged subjectively, that is, indicators that can be clearly recognized based on experience, such as success rate, integration rate, etc. In this case, the weight of the preset first-class indicator can be determined by using the analytic hierarchy process.

[0096] The Analytic Hierarchy Process (AHP) is an algorithm for calculating the importance of different indicators to a given issue. It transforms a qualitative problem into a quantitative one. When different objects involve many evaluation factors, and there are no clear calculation criteria between these factors, it's difficult to determine whether an object is good or bad. The AHP was primarily developed to improve the accuracy of qualitative problems by minimizing subjective comparisons—that is, comparing multiple factors pairwise. The comparison results are then ranked according to importance (i.e., importance scores), resulting in a judgment matrix. To make the results more accurate and authoritative, expert evaluation can be used, where multiple experts contribute their findings. Consistency is then verified through matrix operations on the judgment matrix, and the final weight of each factor is obtained. In essence, the AHP works by using expert judgment to compare a set of indicators pairwise, outputting a quantitative judgment matrix to calculate weights. This method combines psychology and statistics, minimizing subjectivity and achieving quantitative output.

[0097] Specifically, the weight of each preset first-class indicator can be determined by obtaining at least one of the preset first-class indicators and then comparing the results of pairwise comparisons among the at least one preset first-class indicator.

[0098] For example, for the indicators that can be subjectively judged, a scoring matrix can be established by comparing the importance of each pair of indicators to the final conversion. The pairwise comparison process can be evaluated authoritatively using expert judgment. Assuming there are 5 indicators, a 5x5 array matrix can be calculated. The importance score between each pair of indicators can be determined based on the following rules: when two factors are equally important, the scale can be 1; when the former is slightly more important than the latter, the scale can be 3; when the former is significantly more important than the latter, the scale can be 5; when the former is strongly more important than the latter, the scale can be 1; when the former is extremely more important than the latter, the scale can be 9; scales of 2, 4, 6, and 8 can represent the median values ​​of the above adjacent judgments; and if the ratio of the importance of factor i to factor j is a... ij Then the ratio of the importance of factor j to factor i can be expressed as: Other scaling rules are not limited in the embodiments of the present invention.

[0099] For the final score of a certain indicator, the geometric mean of the data set of a single indicator can be used. The importance scores of a certain indicator and all other indicators can be regarded as an array. The geometric mean of the array can be calculated to eliminate the influence of extreme values, and the obtained geometric mean is used as the final score of the indicator. For example, if there are 5 indicators, there are 5 importance scores. When obtaining the final score, the geometric mean of the 5 importance scores can be reduced to one, which is then used as the final score of the indicator. After calculating the geometric mean of each indicator, the final weight of the indicator can be determined according to the proportion of the scores, that is, the proportion of the individual indicator value to the sum of all indicators.

[0100] As another example, if the indicator is a preset second type of indicator, the preset second type of indicator usually refers to some indicators that cannot be judged by experience, that is, indicators whose importance cannot be subjectively defined, such as average order acceptance time, order return rate, etc. In this case, the weight of the preset second type of indicator can be determined by using the entropy weight method.

[0101] The entropy weighting method primarily refers to determining the weight of an indicator by calculating the amount of information it contains. The amount of information contained in an indicator is measured by its information entropy. The greater the amount of information an indicator contains and the lower its information entropy, the greater its weight. Information content is defined as the amount of uncertainty eliminated by an event occurring in a system; its calculation is related to the probability of the event occurring. The higher the probability of an event, the less uncertainty is eliminated after the event occurs, and the lower the amount of information. Information entropy is the average amount of information in all events in the entire system, representing the expected value of each piece of information. In a set of indicator data, the proportion of a single data point to the sum of all data points is used as the probability to calculate the information entropy of that set of data, thus obtaining the indicator's weight. In short, the principle of the entropy weighting method is mainly to analyze existing sample data, calculate the information entropy of each indicator, and then calculate the weight based on the amount of information contained in the indicator.

[0102] Specifically, the weight of each preset second-category indicator can be calculated by obtaining the information entropy of each preset second-category indicator.

[0103] For example, for indicators that cannot be subjectively judged, the proportion of data size can be used as the probability p of the data, the information entropy e of a single indicator data group can be calculated, and then 1-e can be used as the final weight. The formula for calculating the information entropy can be as follows:

[0104]

[0105] It should be noted that different weight ratios can be assigned to the preset first type of indicators and the preset second type of indicators, and the embodiments of the present invention do not impose any restrictions on this.

[0106] In practical applications, when calculating the rank-sum ratio of each pre-assignment object based on the weights of the indicators, if the weights of the indicators are the same, the following formula can be used to calculate the rank-sum ratio:

[0107]

[0108] When the weights of the indicators are not the same, the following formula can be used for calculation:

[0109]

[0110] Among them, R ij It can refer to the rank of the j-th index of the i-th object, W. j This can refer to the weight of the j-th indicator, and the sum of the weights can be 1.

[0111] Step 203: Determine the target order recipient based on the rank-sum ratio of each pre-assigned object.

[0112] The rank-sum ratio (RSR) can be used to indicate the merit level of a pre-assigned candidate. This RSR is primarily positively correlated with the merit level of the pre-assigned candidate. i Or WRSR i The larger the value, the better the evaluation object is.

[0113] In practical applications, pre-assignment objects can be sorted according to their rank-sum ratio. Based on this sorting result, the target order recipient can be determined. The sorting result represents a comprehensive evaluation of the final order recipients. When selecting the target order recipient, the top-ranked pre-assignment object can be chosen as the optimal recipient. This ensures that the optimal recipient achieves the best conversion rate when processing the order.

[0114] In this embodiment of the invention, the rank-sum ratio algorithm is used to comprehensively consider multiple personalized factors of operator product marketing, and based on the ranking results of multiple pre-assigned objects indicated by the rank-sum ratio, it is used to indicate the individual comprehensive evaluation ranking of each pre-assigned object, and to assign an order to the best order-receiving object, so that each assigned order can achieve the best conversion business needs.

[0115] Reference Figure 3 The illustration shows an application scenario diagram of order dispatch provided by an embodiment of the present invention. This application scenario can be a production scenario of actual communication business, involving an order dispatch system 310. The order dispatch system 310 can use a BS architecture and provide services to the outside world through an HTTP interface.

[0116] Specifically, the order dispatch system 310 may include a module 31 for preparing indicator data for order takers, a module 32 for inputting real-time order data, a module 33 for running a comprehensive evaluation model, and a module 34 for outputting the evaluation results of order takers.

[0117] Among them, the order taker indicator data preparation module 31 is mainly responsible for acquiring data such as historical order result data, historical order process data, and order taker attribute data. Since these indicator data require a lot of processing and calculation and are updated regularly, such as once a month, this module can also provide the ability to update indicator data regularly to provide basic data for the calculation of the comprehensive evaluation model.

[0118] The real-time order data input module 32 is mainly responsible for providing an input entry point for real-time pending orders. Real-time order information can include customer location information, order number, etc., to provide a basis for distance calculation and preliminary screening of candidate order takers for the comprehensive evaluation model.

[0119] The comprehensive evaluation model operation module 33 is mainly responsible for using the established comprehensive evaluation model to calculate the comprehensive score of candidate order takers based on the input orders and the order taker indicator data, and outputting the score results and ranking of the candidate order takers. The establishment of the comprehensive evaluation model can be carried out with reference to the method embodiments described in this invention, that is, it can combine the rank-sum ratio method, indicators, and indicator weights to establish an order dispatch model suitable for the telecommunications operator industry. This invention will not elaborate further here. It should be noted that some well-known algorithms used in the creation of this model are publicly available. Other skilled personnel, after understanding the design process of this model, can also create similar comprehensive evaluation models. Comprehensive evaluation models established using the above design process fall within the protection scope of this invention.

[0120] The order taker selection result output module 34 is mainly responsible for combining the input order information and returning the best order taker information, including: employee number, name, organizational structure, etc.

[0121] In this embodiment of the invention, the order allocation system 310 described above can be applied to the production scenario of actual communication services. By calling the established comprehensive evaluation model and using the rank-sum ratio algorithm to comprehensively consider multiple personalized factors of operator product marketing, a multi-dimensional comprehensive evaluation of pre-assignment objects can be achieved. Based on the comprehensive evaluation ranking result for multiple people indicated by the rank-sum ratio, an order is assigned to the best receiving object, ensuring that each assigned order meets the best conversion business needs, thereby improving the overall conversion efficiency of orders. Furthermore, the calculation of indicator weights can be done scientifically, combining qualitative and quantitative methods. For example, the qualitative part can use the analytic hierarchy process (AHP) to reduce errors caused by subjective judgment, while the quantitative part can use the entropy weight method to evaluate weights from the perspective of data information, making the indicators more meaningful. This allows for quantitative analysis of qualitative issues, improving the scientific nature of the evaluation results and reducing errors caused by human subjectivity compared to manually assigning weights. Moreover, by using the rank-sum ratio method of the comprehensive evaluation model to select the most suitable receiving object, the decision factor for order allocation focuses on the ranking of the pre-assignment object rather than the score itself. In addition, the selection of evaluation indicators can be based on three types of data: the order results data of the order recipient, the order marketing process data, and its own data. Evaluation indicators can be mined from the marketing process data and marketing results data and reflected in the comprehensive evaluation results, making the marketing process of the selected order recipients more standardized and regulated.

[0122] It should be noted that, for the sake of simplicity, the method embodiments are all described as a series of actions. However, those skilled in the art should understand that the embodiments of the present invention are not limited to the described order of actions, because according to the embodiments of the present invention, some steps can be performed in other orders or simultaneously. Furthermore, those skilled in the art should also understand that the embodiments described in the specification are preferred embodiments, and the actions involved are not necessarily essential to the embodiments of the present invention.

[0123] Reference Figure 4 The diagram shows a structural block diagram of an embodiment of the order dispatching device of the present invention, which may specifically include the following modules:

[0124] The indicator acquisition module 401 is used to acquire marketing data and obtain at least one indicator from the marketing data for optimizing target orders; wherein the marketing data includes marketing process data and marketing result data;

[0125] The indicator data acquisition module 402 is used to acquire the weight of each indicator and at least one indicator data of at least one pre-assigned object for the at least one indicator.

[0126] The rank-sum ratio generation module 403 is used to generate the rank-sum ratio of each pre-assignment object based on at least one indicator data of the at least one pre-assignment object and the weight of each indicator; the rank-sum ratio is used to indicate the degree of excellence of the pre-assignment object.

[0127] The order receiving object determination module 404 is used to determine the target order receiving object based on the rank-sum ratio of each pre-assigned object.

[0128] In one embodiment of the present invention, the indicator data acquisition module 402 may include the following sub-modules:

[0129] The weight acquisition submodule is used to determine the weight of the preset first type of indicator by using a preset analytic hierarchy process when the indicator is a preset first type of indicator; and / or, to determine the weight of the preset second type of indicator by using a preset entropy weight method when the indicator is a preset second type of indicator.

[0130] In one embodiment of the present invention, the weight acquisition submodule may include the following units:

[0131] The weight acquisition unit is used to acquire at least one of the preset first-class indicators, determine the weight of each preset first-class indicator based on the comparison results of pairs of indicators among the at least one preset first-class indicators, and / or acquire the information entropy of each preset second-class indicator, and calculate the weight of each preset second-class indicator using the information entropy.

[0132] In one embodiment of the present invention, the rank-to-ratio generation module 403 may include the following sub-modules:

[0133] The rank-sum ratio generation submodule is used to obtain the rank of each indicator of each pre-assignment object based on at least one indicator data of the at least one pre-assignment object; and to generate the rank-sum ratio of each pre-assignment object based on the rank of each indicator of each pre-assignment object.

[0134] In one embodiment of the present invention, the rank-to-ratio generation submodule may include the following units:

[0135] A rank generation unit is used to generate a data matrix using the number of metrics for optimizing the target order and the number of pre-assignment objects; wherein each column of the data matrix indicates each metric for optimizing the target order, and each row of the data matrix indicates each pre-assignment object; based on the metric data of each pre-assignment object for each metric, each metric of each pre-assignment object in the data matrix is ​​ranked to obtain a rank matrix; wherein the rank matrix contains the rank of each metric of each pre-assignment object.

[0136] In one embodiment of the present invention, the rank-sum ratio is positively correlated with the evaluation level of the pre-assigned object; the order-receiving object determination module 404 may include the following sub-modules:

[0137] The order receiving object determination submodule is used to sort each pre-assigned object according to the rank-sum ratio of each pre-assigned object; and determine the target order receiving object based on the sorting result.

[0138] In this embodiment of the invention, the order allocation device obtains at least one indicator for optimizing target orders from marketing process data and marketing result data, and obtains the weight of each indicator. Then, using at least one pre-assignment object, and based on at least one indicator data and the weight of each indicator, the rank-sum ratio of each pre-assignment object is obtained. This rank-sum ratio is mainly used to indicate the evaluation level of the pre-assignment object. At this time, the target order receiving object to be assigned to the target order can be determined based on the rank-sum ratio of each pre-assignment object. By comprehensively considering multiple personalized factors of operator product marketing through the rank-sum ratio algorithm, and based on the ranking result of multiple pre-assignment objects indicated by the rank-sum ratio, it is used to indicate the individual comprehensive evaluation ranking of each pre-assignment object, and to assign a certain order to the best receiving object, so that each assigned order can achieve the best conversion business needs.

[0139] As the device embodiment is basically similar to the method embodiment, the description is relatively simple, and relevant parts can be found in the description of the method embodiment.

[0140] This invention also provides an electronic device, comprising:

[0141] It includes a processor, a memory, and a computer program stored in the memory and capable of running on the processor. When the computer program is executed by the processor, it implements the various processes of the above-described order dispatch method embodiments and achieves the same technical effect. To avoid repetition, it will not be described again here.

[0142] This invention also provides a computer-readable storage medium storing a computer program. When the computer program is executed by a processor, it implements the various processes of the above-described order dispatch method embodiments and achieves the same technical effect. To avoid repetition, it will not be described again here.

[0143] The various embodiments in this specification are described in a progressive manner, with each embodiment focusing on the differences from other embodiments. The same or similar parts between the various embodiments can be referred to each other.

[0144] Those skilled in the art will understand that embodiments of the present invention can be provided as methods, apparatus, or computer program products. Therefore, embodiments of the present invention can take the form of entirely hardware embodiments, entirely software embodiments, or embodiments combining software and hardware aspects. Furthermore, embodiments of the present invention can take the form of computer program products implemented on one or more computer-usable storage media (including but not limited to disk storage, CD-ROM, optical storage, etc.) containing computer-usable program code.

[0145] This invention is described with reference to flowchart illustrations and / or block diagrams of methods, terminal devices (systems), and computer program products according to embodiments of the invention. It will be understood that each block of the flowchart illustrations and / or block diagrams, and combinations of blocks in the flowchart illustrations and / or block diagrams, can be implemented by computer program instructions. These computer program instructions can be provided to a processor of a general-purpose computer, special-purpose computer, embedded processor, or other programmable data processing terminal device to produce a machine, such that the instructions, which execute via the processor of the computer or other programmable data processing terminal device, generate instructions for implementing the flowchart illustrations and / or block diagrams. Figure 1 One or more processes and / or boxes Figure 1 A device that provides the functions specified in one or more boxes.

[0146] These computer program instructions may also be stored in a computer-readable storage medium that can direct a computer or other programmable data processing terminal device to operate in a particular manner, such that the instructions stored in the computer-readable storage medium produce an article of manufacture including instruction means, which are implemented in a process Figure 1 One or more processes and / or boxes Figure 1The function specified in one or more boxes.

[0147] These computer program instructions can also be loaded onto a computer or other programmable data processing terminal equipment, causing a series of operational steps to be performed on the computer or other programmable terminal equipment to produce a computer-implemented process, thereby providing instructions that execute on the computer or other programmable terminal equipment for implementing the process. Figure 1 One or more processes and / or boxes Figure 1 The steps of the function specified in one or more boxes.

[0148] Although preferred embodiments of the present invention have been described, those skilled in the art, upon learning the basic inventive concept, can make other changes and modifications to these embodiments. Therefore, the appended claims are intended to be interpreted as including the preferred embodiments as well as all changes and modifications falling within the scope of the embodiments of the present invention.

[0149] Finally, it should be noted that the user information (including but not limited to user device information, user personal information, etc.) and data (including but not limited to data used for analysis, data stored, data displayed, etc.) involved in this application are all information and data authorized by the user or fully authorized by all parties. Furthermore, the collection, use and processing of the relevant data must comply with the relevant laws, regulations and standards of the relevant countries and regions, and corresponding operation portals are provided for users to choose to authorize or refuse.

[0150] It should also be noted that, in this document, relational terms such as "first" and "second" are used only to distinguish one entity or operation from another, and do not necessarily require or imply any such actual relationship or order between these entities or operations. Furthermore, the terms "comprising," "including," or any other variations thereof are intended to cover non-exclusive inclusion, such that a process, method, article, or terminal device that comprises a list of elements includes not only those elements but also other elements not expressly listed, or elements inherent to such a process, method, article, or terminal device. Without further limitations, an element defined by the phrase "comprising one..." does not exclude the presence of other identical elements in the process, method, article, or terminal device that includes said element.

[0151] The above provides a detailed description of an order dispatch method, an order dispatch device, a corresponding electronic device, and a corresponding computer-readable storage medium provided by the present invention. Specific examples have been used to illustrate the principles and implementation methods of the present invention. The description of the above embodiments is only for the purpose of helping to understand the method and core ideas of the present invention. At the same time, for those skilled in the art, there will be changes in the specific implementation methods and application scope based on the ideas of the present invention. Therefore, the content of this specification should not be construed as a limitation of the present invention.

Claims

1. An order dispatch method, characterized in that, The method includes: Obtain marketing data and attribute data of historical orders for pre-assigned objects, and extract at least one indicator from the marketing data and attribute data for optimizing target orders; wherein, the marketing data includes marketing process data and marketing result data; the marketing result data includes: success rate, integration rate, customer value, and service satisfaction; the marketing process data includes second-party completion rate, average order acceptance time, return rate, first-contact time, average return time, and order acceptance rate; the attribute data includes distance; Obtaining the weights of each indicator and at least one indicator data for at least one pre-assigned object for the at least one indicator includes: if the indicator is a preset first type of indicator, then using a preset analytic hierarchy process to determine the weight of the preset first type of indicator, wherein the preset first type of indicator is an indicator whose relative importance is determined subjectively; and / or, if the indicator is a preset second type of indicator, then using a preset entropy weight method to determine the weight of the preset second type of indicator, wherein the preset second type of indicator is an indicator whose importance cannot be subjectively determined; the weights are used to represent the relative importance of the indicator. Based on at least one indicator data and the weight of each indicator of the at least one pre-assigned object, the rank-sum ratio of each pre-assigned object is generated, including: obtaining the rank of each indicator of each pre-assigned object based on at least one indicator data of the at least one pre-assigned object; generating the rank-sum ratio of each pre-assigned object based on the rank of each indicator of each pre-assigned object and the weight of each indicator; the rank-sum ratio is used to indicate the degree of excellence of the pre-assigned object. The target order recipient is determined based on the rank-sum ratio of each pre-assigned object; The step of obtaining the rank of each indicator of each pre-assignment object based on at least one indicator data of the at least one pre-assignment object includes: generating a data matrix using the number of indicators for optimizing the target order and the number of pre-assignment objects; wherein each column of the data matrix indicates each indicator for optimizing the target order, and each row of the data matrix indicates each pre-assignment object; ranking each indicator of each pre-assignment object in the data matrix according to the indicator data of each pre-assignment object for each indicator, to obtain a rank matrix; wherein the rank matrix contains the rank of each indicator of each pre-assignment object.

2. The method according to claim 1, characterized in that, The step of determining the weights of the preset first-type indicators using a preset analytic hierarchy process includes: Obtain at least one of the preset first-category indicators; The weights of each preset first-category indicator are determined based on the comparison results of at least one pairwise indicator among the preset first-category indicators.

3. The method according to claim 1 or 2, characterized in that, The step of determining the weights of the preset second type of index using the preset entropy weight method includes: Obtain the information entropy of each preset second-type indicator, and use the information entropy to calculate the weight of each preset second-type indicator.

4. The method according to claim 1, characterized in that, The rank-sum ratio is positively correlated with the evaluation level of the pre-assigned objects; the step of determining the target order recipients based on the rank-sum ratio of each pre-assigned object includes: Sort each pre-assignment object according to the rank-sum ratio of each pre-assignment object; Based on the sorting results, the target order recipients are determined.

5. An order dispatching device, characterized in that, The device includes: The indicator acquisition module is used to acquire marketing data and attribute data of historical orders of pre-assigned objects, and to extract at least one indicator for optimizing target orders from the marketing data and attribute data; wherein, the marketing data includes marketing process data and marketing result data; the marketing result data includes: success rate, integration rate, customer value, and service satisfaction; the marketing process data includes second-party completion rate, average order acceptance time, return rate, first-contact time, average return time, and order acceptance rate; the attribute data includes distance; The indicator data acquisition module is used to acquire the weights of each indicator and at least one indicator data for at least one pre-assigned object for the at least one indicator. It includes: a weight acquisition submodule, used to determine the weights of the preset first-type indicators using a preset analytic hierarchy process when the indicator is a preset first-type indicator, wherein the preset first-type indicators are indicators whose relative importance is subjectively judged; and / or, when the indicator is a preset second-type indicator, used to determine the weights of the preset second-type indicators using a preset entropy weight method, wherein the preset second-type indicators are indicators whose importance cannot be subjectively defined; the weights are used to represent the relative importance of the indicator. A rank-sum ratio generation module is used to generate the rank-sum ratio of each pre-assignment object based on at least one indicator data and the weight of each indicator of the at least one pre-assignment object. This includes: a rank-sum ratio generation submodule, used to obtain the rank of each indicator of each pre-assignment object based on at least one indicator data of the at least one pre-assignment object; and to generate the rank-sum ratio of each pre-assignment object based on the rank of each indicator and the weight of each indicator. The rank-sum ratio is used to indicate the degree of excellence of the pre-assignment object. The rank-sum ratio generation submodule includes: a rank generation unit, used to generate a data matrix using the number of indicators for optimizing the target order and the number of pre-assignment objects; wherein each column of the data matrix indicates each indicator for optimizing the target order, and each row of the data matrix indicates each pre-assignment object; and to rank each indicator of each pre-assignment object in the data matrix based on the indicator data of each pre-assignment object for each indicator, obtaining a rank matrix; wherein the rank matrix contains the rank of each indicator of each pre-assignment object. The order receiving object determination module is used to determine the target order receiving object based on the rank-sum ratio of each pre-assigned object.

6. An electronic device, characterized in that, include: A processor, a memory, and a computer program stored in the memory and capable of running on the processor, wherein the computer program, when executed by the processor, implements the order dispatch method as described in any one of claims 1 to 4.

7. A computer-readable storage medium, characterized in that, A computer program is stored on the computer-readable storage medium, which, when executed by a processor, implements the order dispatch method as described in any one of claims 1 to 4.

Images