Medical supply chain traceability system and method based on association rule analysis

Abstract

The invention discloses a medical supply chain traceability system and method based on association rule analysis, and relates to the technical field of medical consumable supply chain management.The specification weight is calculated according to the number of times of use and a fluctuation coefficient, the use frequency is calculated according to the unit cycle time length, and then a specification concentrated reference value is calculated through the specification weight and the use frequency; calculating a dispersity factor according to the demand distribution factor and the average use time interval, calculating the use frequency according to the number of first selection times and the unit cycle time length, finally calculating a specification dispersity reference value according to the dispersity factor and the use frequency, obtaining a matching index of a selected division value in the sequence, carrying out weighted operation to obtain a matching index threshold value, and finally calculating the index value according to the matching index threshold value. And determining a distribution strategy of the target consumable category and a target time period, obtaining a use record of the medical consumables in the target time period, and performing time sequence evaluation on the specification concentration characteristics and the specification dispersion characteristics of the medical consumables.

Description

Technical Field

[0001] The present invention relates to the technical field of medical consumables supply chain management, and specifically to a medical supply chain traceability system and method based on association rule analysis. Background Art

[0002] Traditional medical consumables inventory management methods often rely on experience or simple statistical data, lacking in-depth analysis of the usage patterns of medical consumables, resulting in unreasonable inventory levels, which in turn put pressure on the procurement, inventory, and logistics links of the supply chain, potentially leading to inventory backlogs or shortages in the supply chain and increasing supply chain costs.

[0003] In addition, existing medical consumables management systems often lack an effective early warning mechanism and are unable to detect problems such as inventory shortages and abnormal usage in a timely manner. This makes it difficult to identify and respond to risks in the supply chain in a timely manner, increasing the risk of supply chain disruptions. At the same time, traditional medical consumables rationing strategies are often fixed and cannot be dynamically adjusted according to changes in actual demand, resulting in insufficient supply of consumables during peak demand periods and overstocking during low demand periods. This makes it difficult for the supply chain to adapt to demand changes and reduces the flexibility and responsiveness of the supply chain. Summary of the Invention

[0004] The purpose of the present invention is to provide a medical supply chain traceability system and method based on association rule analysis to solve the problems raised in the prior art.

[0005] To achieve the above objective, the present invention provides the following technical solution: A medical supply chain traceability method based on association rule analysis, the method comprising: Step S100: Determine a certain type of medical consumable as the target consumable and obtain the usage records of all specifications within a reference time range; Step S200: Calculate the specification weight based on the usage times and the fluctuation coefficient, calculate the usage frequency according to the unit cycle time length, and then calculate the reference value in the specification set through the specification weight and the usage frequency; Step S300: Calculate the demand distribution factor by combining the first selection times and the total selection times, calculate the dispersion factor based on the demand distribution factor and the average usage time interval, calculate the usage frequency again according to the first selection times and the unit cycle time length, and finally calculate the specification dispersion reference value from the dispersion factor and the usage frequency; Step S400: According to the reference value in the specification set and the dispersion reference value, calculate the matching index for each unit cycle in combination with a given weight and sort them in ascending order, obtain the matching index of the selected division value in the sequence, and perform weighted operation to obtain the matching index threshold; Step S500: Determine the rationing strategy and target time period for the target consumable category, obtain the usage records of medical consumables during the target time period, and conduct a time series assessment of the concentrated characteristics and dispersed characteristics of the specifications of medical consumables; Step S600: Perform a weighted operation on the characteristic values corresponding to the concentrated characteristics and dispersed characteristics of the specifications to obtain a consumption characteristic evaluation value. When the difference from the matching index threshold exceeds the difference threshold, an alarm prompt is sent to the management personnel.

[0006] Further, step S100 includes: Step S101: Take a certain type of medical consumable as the target consumable category, and obtain the usage records of all specifications of the target consumable within the reference time range; Step S102: Divide the reference time range into several unit reference periods with a time length of T1, and obtain all the usage records of the k-th specification consumable in the target category during the c-th unit reference period in the reference time range.

[0007] Further, step S200 includes: Step S201: In the c-th unit reference period, obtain the total number of usage times Q of the medical consumables corresponding to the target consumable category v,c and the number of usage times Q of the k-th specification consumable k,c , divide the c-th unit reference period into i interval segments, and calculate the proportion of the number of usage times of the k-th specification consumable in each interval segment to the total number of usage times of the medical consumables corresponding to the target consumable category; Step S202: Collect the data values of the proportions of all i interval segments, and record the variance of the data values of the proportions as the fluctuation coefficient D k,c ; Step S203: Calculate the specification weight W of the k-th specification consumable in the c-th unit reference period k,c , ; Step S204: Calculate the usage frequency F of the k-th specification consumable in the c-th unit reference period k,c , F k,c =Q k,c / tc, where tc represents the time length of the c-th unit reference period; Step S205: Calculate the concentrated reference value A of the specifications of the medical consumables corresponding to the target consumable category in the c-th unit reference period c , , where n represents the total number of specifications of the medical consumables in the target category.

[0008] Further, step S300 includes: Step S301: Obtain the usage scenarios of the medical consumables corresponding to the target consumable category from the usage records of the target consumable category. The usage scenarios include intensive care, emergency, general outpatient clinic, and chronic disease outpatient clinic, and set the weight factor for each usage scenario; Step S302: Obtain the application scenario of each use of the k-th specification consumable in the c-th unit reference period, and use the average value of the weight factors as the priority coefficient P of the k-th specification consumable in the c-th unit reference period k,c ; Step S303: In the c-th unit reference period, obtain the number of times N that the k-th specification consumable is first selected, and the total number of times N that the medical consumables corresponding to the target consumable category are selected k,c , and calculate the demand distribution factor R of the c-th unit reference period v,c , k,c , ; Step S304: In the c-th unit reference period, obtain the average usage time interval Lk,c of the k-th specification consumable, and calculate the dispersion factor V of the k-th specification consumable in the c-th unit reference period k,c , , where R j,c represents the demand distribution factor of the j-th specification consumable calculated in the c-th unit reference period; Step S305: Calculate the usage frequency E of the k-th specification consumable in the c-th unit reference period k,c , E k,c =N k,c / tc; Step S306: Calculate the specification dispersion reference value D of the medical consumables corresponding to the target consumable category in the c-th unit reference period c , .

[0009] Further, step S400 includes: Step S401: Calculate the matching index M of the c-th unit reference period c , M c =ω1×A c +ω2×D c , where ω1 and ω2 represent weights, satisfying the conditions ω1 > 0, ω2 > 0 and ω1 + ω2 = 1; Step S402: Obtain the matching indexes of M unit periods, and arrange all M matching indexes in ascending order according to the values to obtain a matching index sequence; Step S403: Set the balance coefficient α, satisfying the condition 0.25 ≤ α ≤ 0.75, and calculate the position serial number x in the sequence, x = α × (M + 1); Step S404: When x is an integer, use the x-th matching exponent in the matching exponent sequence as the matching exponent threshold Mt. When x is not an integer, round x up and down to obtain two integers h and h + 1; Step S405: Use the integer with the smallest absolute value of the difference from x between h and h + 1 as the reference value h0, and calculate the matching exponent threshold Mt, where Mt = M h0 + α × (M h+1 - M h ).

[0010] By dividing the reference time range into multiple unit reference periods and calculating key metrics such as the usage times and fluctuation coefficients of each specification of consumables within each period, this method can capture the periodic changes in consumable usage. This dynamic analysis method makes the adjustment of inventory management and rationing strategies more scientific and timely, and helps to further improve resource utilization efficiency.

[0011] Furthermore, step S500 includes: Step S501: Obtain the rationing strategy for the target consumable category in the target time period, use the rationing strategy as the target rationing strategy, and use the time period with the same rationing strategy for a segment of the target consumable category as the target time period; Step S502: Divide the target time period into several target unit periods with a time length of T2, where T2 < T1. Obtain all the usage records of the k-th specification of consumables in the target category in the d-th target unit period of the target time period; In the monitoring of the target unit period, to improve the monitoring efficiency and promptly detect the rationing risks of medical consumables, a shorter monitoring period is used. However, a shorter monitoring period is prone to being affected by accidental events and generating errors. Therefore, in short-term monitoring, the collection of continuous cycle dependence characteristics is strengthened, making the observed values of the target unit period more accurate; Step S503: Calculate the specification weight W k,d of the k-th specification of consumables in the target consumable category in the d-th target unit period, obtain the usage frequency F k,d of the k-th specification of consumables in the d-th target unit period, and calculate the centralized reference value B d of the specifications of the medical consumables corresponding to the target consumable category in the d-th target unit period, ; Step S504: Calculate the temporal feature value S d+1 of the (d + 1)-th specification set, , where λ represents a coefficient satisfying the condition 0 < λ < 1.

[0012] Furthermore, step S500 includes: Step S505: Obtain the application scenarios of the consumables of the k-th specification each time they are used in the d-th target unit cycle, and calculate the demand distribution factor R of the c-th unit reference cycle k,d ; Step S506: Obtain the average usage time interval L of the consumables of the k-th specification in the d-th target unit cycle k,d , calculate the dispersion factor V of the consumables of the k-th specification in the d-th target unit cycle k,d , and obtain the usage frequency E of the consumables of the k-th specification in the d-th target unit cycle k,d ; Step S507: Calculate the dispersion reference value G of the medical consumables corresponding to the target consumable category in the d-th target unit cycle d , ; Step S508: Calculate the temporal characteristic value U of the (d + 1)-th specification dispersion characteristic d+1 , .

[0013] By continuously monitoring and evaluating the consumption characteristics, this method can timely detect potential problems, such as abnormal inventory, demand changes, etc., and issue warning prompts in advance. This provides sufficient reaction time for managers to quickly take measures to deal with potential risks and reduce losses.

[0014] Furthermore, step S600 includes: Step S601: Calculate the (d + 1)-th consumption characteristic evaluation value γ d+1 , γ d+1 = ω1×S d+1 + ω2×U d+1 ; Step S602: Set the difference threshold β. When |Mt - γ d+1 | > β, issue a warning prompt to the manager of the target consumable category.

[0015] To better implement the above method, a pharmaceutical supply chain traceability system based on association rule analysis is also proposed. The system includes: a historical record management module, a historical feature management module, a feature comparison module, and a difference evaluation module; The historical record management module is used to store the usage records of medical consumables. The historical feature management module is used to collect the specification centralized reference value and the specification dispersion reference value in the usage records of medical consumables, and calculate the matching index threshold. The feature comparison module is used to compare the consumption characteristics of medical consumables in the target time period and the reference time range. The difference evaluation module is used to evaluate the degree of difference.

[0016] Further, the historical feature management module includes a specification weight management unit, a centralized reference value calculation unit, a demand distribution factor management unit, a dispersion factor management unit, a dispersion reference value calculation unit, a matching index management unit, and a matching index threshold management unit; among them, the specification weight management unit is used to calculate the specification weight of a certain specification of medical consumables, the centralized reference value calculation unit is used to calculate the centralized reference value of the medical consumables, the demand distribution factor management unit is used to calculate the demand distribution factor of a certain specification of medical consumables, the dispersion factor management unit is used to calculate the dispersion factor of a certain specification of medical consumables, the dispersion reference value calculation unit is used to calculate the specification dispersion reference value of the medical consumables, the matching index management unit is used to calculate the matching index of a unit period, and the matching index threshold management unit is used to calculate the matching index threshold; Further, the feature comparison module includes a feature value management unit and a timing feature management unit; among them, the feature value management unit is used to obtain the specification weight, centralized reference value, demand distribution factor, dispersion factor, and specification dispersion reference value of the target consumable category in the target time period, and the timing feature management unit is used to calculate the timing feature value of the centralized specification and the timing feature value of the dispersed specification; Further, the difference evaluation module includes: a consumption feature management unit and an information reminder unit; among them, the consumption feature management unit is used to calculate the consumption feature evaluation value, and the information reminder unit is used to manage the difference threshold and give an information prompt for the judgment result that meets the alarm condition.

[0017] Compared with the prior art, the beneficial effects of the present invention are as follows: By determining the target consumable category and obtaining the usage records, the present invention provides a data basis for subsequent analysis; by dividing the time range and calculating the weights and indicators, it more accurately reflects the usage rules of medical consumables; by calculating the matching index threshold, it provides a scientific basis for formulating subsequent allocation strategies; by evaluating the centralized and dispersed features of the specifications, it timely discovers potential problems and issues alarm prompts; finally, based on the above analysis results, a scientific and reasonable medical consumable allocation strategy is formulated. By deeply analyzing the usage rules of medical consumables, combining scientific calculation methods and early warning mechanisms, this method not only optimizes the inventory management, allocation efficiency, and risk warning capabilities of medical consumables, but also improves the overall efficiency and response capabilities of the supply chain, providing strong support for the reasonable allocation and management of medical consumables and the optimization of the supply chain. BRIEF DESCRIPTION OF THE DRAWINGS

[0018] Figure 1 It is a schematic structural diagram of the medical supply chain traceability system based on association rule analysis of the present invention; Figure 2 It is a schematic flow diagram of the medical supply chain traceability method based on association rule analysis of the present invention. DETAILED DESCRIPTION OF THE EMBODIMENTS

[0019] Next, the technical solutions in the embodiments of the present invention will be clearly and completely described in conjunction with the accompanying drawings in the embodiments of the present invention. Obviously, the described embodiments are only a part of the embodiments of the present invention, rather than all the embodiments. All other embodiments obtained by those of ordinary skill in the art based on the embodiments of the present invention without creative efforts shall fall within the protection scope of the present invention.

[0020] Embodiment: As Figure 1 - Figure 2 shown, the present invention provides a technical solution, a medical supply chain traceability system and method based on association rule analysis; The method includes: Step S100: Determine a certain type of medical consumables as the target consumables, and obtain the usage records of all specifications within the reference time range; Among them, step S100 includes: Step S101: Take a certain type of medical consumables as the target consumables category, and obtain the usage records of all specifications of the target consumables within the reference time range; Step S102: Divide the reference time range into several unit reference periods with a time length of T1, and obtain all the usage records of the k-th specification consumables in the target category in the c-th unit reference period in the reference time range.

[0021] Step S200: Calculate the specification weight according to the usage times and the fluctuation coefficient, calculate the usage frequency according to the unit cycle time length, and then calculate the specification centralized reference value through the specification weight and the usage frequency; Among them, step S200 includes: Step S201: In the c-th unit reference period, obtain all the usage times Q v,c of the medical consumables corresponding to the target consumables category k,c and the usage times Q of the k-th specification consumables, divide the c-th unit reference period into i interval segments, and calculate the proportion of the usage times of the k-th specification consumables in each interval segment to all the usage times of the medical consumables corresponding to the target consumables category; k,c ; Step S202: Collect the data values of the proportions of all i interval segments, and record the variance of the data values of the proportions as the fluctuation coefficient D k,c ; Step S202: Collect the data values of the proportions of all i interval segments, and record the variance of the data values of the proportions as the fluctuation coefficient D k,c ; ; Step S203: Calculate the specification weight W k,c of the k-th specification consumables in the c-th unit reference period ; Step S203: Calculate the specification weight W k,c of the k-th specification consumables in the c-th unit reference period, F k,c =Q k,c / tc, where tc represents the time length of the c-th unit reference period; Step S205: Calculate the centralized reference value A of the medical consumables corresponding to the target consumable category in the c-th unit reference period c , , where n represents the total number of specifications of the medical consumables in the target category.

[0022] Step S300: Calculate the demand distribution factor by combining the first selection times and the total selection times, calculate the dispersion factor based on the demand distribution factor and the average usage time interval, calculate the usage frequency again according to the first selection times and the unit cycle time length, and finally calculate the specification dispersion reference value from the dispersion factor and the usage frequency; Among them, Step S300 includes: Step S301: Obtain the usage scenarios of the medical consumables corresponding to the target consumable category from the usage records of the target consumable category. The usage scenarios include intensive care, emergency, general outpatient clinic, and chronic disease outpatient clinic, and set the weight factors for each usage scenario; Step S302: Obtain the application scenarios of the k-th specification consumables each time they are used in the c-th unit reference period, and take the average value of the weight factors as the priority coefficient P of the k-th specification consumables in the c-th unit reference period k,c ; Step S303: In the c-th unit reference period, obtain the number of times N that the k-th specification consumables are first selected k,c , the total number of times N that the medical consumables corresponding to the target consumable category are selected v,c , calculate the demand distribution factor R of the c-th unit reference period k,c , ; Step S304: In the c-th unit reference period, obtain the average usage time interval Lk,c of the k-th specification consumables, and calculate the dispersion factor V of the k-th specification consumables in the c-th unit reference period k,c , , where R j,c represents calculating the demand distribution factor of the j-th specification consumables in the c-th unit reference period; Step S305: Calculate the usage frequency E of the k-th specification consumables in the c-th unit reference period k,c , E k,c =N k,c / tc; Step S306: Calculate the specification dispersion reference value D of the medical consumables corresponding to the target consumable category in the c-th unit reference period c , .

[0023] Step S400: Calculate the matching index for each unit period according to the reference value and dispersion reference value in the specification set, combined with the given weights, and arrange them in ascending order. Obtain the matching index of the selected graduation value in the sequence, and perform weighted operation to obtain the matching index threshold; Among them, step S400 includes: Step S401: Calculate the matching index M of the c-th unit reference period c , M c = ω1×A c + ω2×D c , where ω1 and ω2 represent weights, satisfying the conditions ω1 > 0, ω2 > 0 and ω1 + ω2 = 1; Step S402: Obtain the matching indices of M unit periods, and arrange all M matching indices in ascending order according to the numerical values to obtain the matching index sequence; Step S403: Set the balance coefficient α, satisfying the condition 0.25 ≤ α ≤ 0.75, calculate the position serial number x in the sequence, x = α×(M + 1); Step S404: When x is an integer, take the x-th matching index in the matching index sequence as the matching index threshold Mt. When x is not an integer, round x up and down to obtain two integers h and h + 1; Step S405: Take the integer with the smallest absolute value of the difference from x among h and h + 1 as the reference value h0, and calculate the matching index threshold Mt, Mt = M h0 + α×(M h+1 - M h ).

[0024] Step S500: Determine the rationing strategy and target time period of the target consumable category, obtain the usage records of medical consumables in the target time period, and perform a time series evaluation on the specification concentration characteristics and specification dispersion characteristics in the specification set of medical consumables; Among them, step S500 includes: Step S501: Obtain the rationing strategy of the target consumable category in the target time period, take the rationing strategy as the target rationing strategy, and take a time period with the same rationing strategy as the target consumable category as the target time period; Step S502: Divide the target time period into several target unit periods with a time length of T2, where T2 < T1. Obtain all the usage records of the k-th specification consumable in the target category in the d-th target unit period of the target time period; Step S503: Calculate the specification weight W of the k-th specification consumable in the target consumable category in the d-th target unit period k,d , and obtain the usage frequency F of the k-th specification consumable in the d-th target unit period k,d, calculate the reference value B of the specification set of the medical consumables corresponding to the target consumable category in the d-th target unit period d , ; Step S504: Calculate the temporal feature value S of the (d + 1)-th specification set d+1 , , where λ represents a coefficient, satisfying the condition 0 < λ < 1; Step S505: Obtain the application scenarios of each use of the k-th specification of consumables in the d-th target unit period, and calculate the demand distribution factor R of the c-th unit reference period k,d ; Step S506: Obtain the average usage time interval L of the k-th specification of consumables in the d-th target unit period k,d , calculate the dispersion factor V of the k-th specification of consumables in the d-th target unit period k,d , and obtain the usage frequency E of the k-th specification of consumables in the d-th target unit period k,d ; Step S507 Calculate the dispersion reference value G of the medical consumables corresponding to the target consumable category in the d-th target unit period d , ; Step S508 Calculate the temporal feature value U of the (d + 1)-th specification dispersion feature d+1 , .

[0025] Step S600: Perform a weighted operation on the feature values corresponding to the specification set feature and the specification dispersion feature to obtain a consumption feature evaluation value. When the difference from the matching index threshold exceeds the difference threshold, an alarm prompt is sent to the management personnel; Among them, Step S600 includes: Step S601: Calculate the (d + 1)-th consumption feature evaluation value γ d+1 , γ d+1 = ω1 × S d+1 + ω2 × U d+1 ; Step S602: Set the difference threshold β. When |Mt - γ d+1 | > β, an alarm prompt is sent to the management personnel of the target consumable category.

[0026] Example:

[0027] A certain hospital selects "infusion set" as the target consumable category and collects the usage records of different specifications of infusion sets. In the example, there are 3 specifications of syringes, and at this time n = 3, namely ordinary type, precision filter type, and light-shielding type.

[0028] In the embodiment, c = 1, T1 = 10 days, and the usage record Q of the target consumable category is obtained. v,1 = 150 times, where the ordinary type Q 1,1 = 60 times, the precision filtration type Q 2,1 = 50 times, the light-proof type Q 3,1 = 40 times; When i = 5, these 10 days are divided into 5 interval segments. The usage frequency ratios of the ordinary type in each interval segment are 0.3, 0.4, 0.3, 0.2, and 0.2 respectively. According to the variance formula, D 1,1 = 0.004 is obtained. Similarly, the fluctuation coefficients D of the precision filtration type and the light-proof type are obtained. 2，1 = 0.006, D 3,1 = 0.005.

[0029] According to the formula , the specification weight W of the ordinary type is calculated. 1,1 , the specification weight W of the precision filtration type 2,1 , and the specification weight W of the light-proof type 3,1 ; , , ; The usage frequency F is calculated. 1,1 = 60 / 10 = 6, F 2,1 = 50 / 10 = 5, F 3,1 = 40 / 10 = 4; According to the formula , A1 = 5.11 is calculated.

[0030] The usage scenarios and weight factors are set as follows: intensive care 0.9, emergency 0.7, general outpatient 0.4, chronic disease outpatient 0.3; In the 10-day interval, the ordinary type infusion set is used for intensive care 5 times, emergency 10 times, general outpatient 45 times, and chronic disease outpatient 0 times. Therefore, the priority coefficient P of the ordinary type infusion set 1，1 = 0.44. At the same time, the precision filtration type P is obtained. 2,1 = 0.47, and the light-proof type P 3,1 = 0.38; The number of times the ordinary type is selected for the first time N 1,1 = 20, N 2,1 = 80, N 3,1 = 50. According to the formula , R is calculated. 1，1 = 0.021. At the same time, the precision filtration type R is obtained. 2,1 = 0.089, and the light-proof type R 3,1 = 0.047, ΣR j，1= 0.157; According to the formula calculate V 1,1 = 0.113, V 2,1 = 0.4646, V 3,1 = 0.2; Therefore, D1 = 2 × 0.113 + 8 × 0.464 + 5 × 0.2 = 4.938; Set the weight ω1 = 0.7, ω2 = 0.3, M1 = ω1 × A1 + ω2 × D1 = 5.0584; Obtain 5 unit cycle matching indices, sort them in ascending order, [4.8956, 4.9758, 5.0584, 5.1725, 5.2032]; Set α = 0.4, x = 0.4 × (5 + 1) = 2.4, so obtain M2 = 4.9758 and M3 = 5.0584 for calculation, Mt = 4.9758 + 0.4 × (5.0584 - 4.9758) = 5.00884 ≈ 5.

[0031] In the embodiment, S d = 5.22, B d+1 = 5.37, U d = 4.87, G d+1 = 5.21, λ = 0.7; Calculate to obtain S d+1 = 5.325, U d+1 = 5.108, γ d+1 = 0.7 × 5.325 + 0.3 × 5.108 = 5.26; Set the difference threshold β = 0.5, so no alarm is made.

[0032] The system includes: a historical record management module, a historical feature management module, a feature comparison module, and a difference evaluation module; Among them, the historical record management module is used to store the usage records of medical consumables.

[0033] Among them, the historical feature management module is used to collect the reference value of the specification set and the reference value of the specification dispersion degree in the usage record of medical consumables, and calculate the matching index threshold. Among them, the historical feature management module includes a specification weight management unit, a centralized reference value calculation unit, a demand distribution factor management unit, a dispersion factor management unit, a dispersion reference value calculation unit, a matching index management unit, and a matching index threshold management unit; among them, the specification weight management unit is used to calculate the specification weight of a certain specification of medical consumables, the centralized reference value calculation unit is used to calculate the centralized reference value of the specification of medical consumables, the demand distribution factor management unit is used to calculate the demand distribution factor of a certain specification of medical consumables, the dispersion factor management unit is used to calculate the dispersion factor of a certain specification of medical consumables, the dispersion reference value calculation unit is used to calculate the reference value of the specification dispersion degree of medical consumables, the matching index management unit is used to calculate the matching index of a unit cycle, and the matching index threshold management unit is used to calculate the matching index threshold.

[0034] Among them, the feature comparison module is used to compare the consumption characteristics of medical consumables in the target time period and the reference time range. Among them, the feature comparison module includes a feature value management unit and a timing feature management unit; among them, the feature value management unit is used to obtain the specification weight, the centralized reference value of the specification, the demand distribution factor, the dispersion factor, and the reference value of the specification dispersion degree of the target consumable category in the target time period, and the timing feature management unit is used to calculate the timing feature value of the centralized specification and the timing feature value of the dispersed specification.

[0035] Among them, the difference evaluation module is used to evaluate the degree of difference. Among them, the difference evaluation module includes: a consumption feature management unit and an information reminder unit; among them, the consumption feature management unit is used to calculate the consumption feature evaluation value, and the information reminder unit is used to manage the difference threshold and perform information prompts on the determination results that meet the alarm conditions.

[0036] For those skilled in the art, it is obvious that the present invention is not limited to the details of the above exemplary embodiments, and can be implemented in other specific forms without departing from the spirit or basic characteristics of the present invention. Therefore, from any point of view, the embodiments should be regarded as exemplary and non-limiting. The scope of the present invention is defined by the appended claims rather than the above description. Therefore, it is intended to include all changes falling within the meaning and scope of the equivalent elements of the claims in the present invention.

Claims

1. A traceability method for the pharmaceutical supply chain based on association rule analysis, characterized in that: The method includes the steps of: Step S100: Determine a certain type of medical consumables as the target consumables, and obtain the usage records of all specifications within the reference time range; Step S200: Calculate the specification weight according to the usage times and the fluctuation coefficient, calculate the usage frequency according to the unit cycle time length, and then calculate the reference value in the specification set through the specification weight and the usage frequency; Step S300: Calculate the demand distribution factor by combining the first selection times and the total selection times, calculate the dispersion factor according to the demand distribution factor and the average usage time interval, calculate the usage frequency again according to the first selection times and the unit cycle time length, and finally calculate the specification dispersion reference value by the dispersion factor and the usage frequency; Step S400: According to the reference value in the specification set and the dispersion reference value, calculate the matching index of each unit cycle by combining the given weight and arrange them in ascending order, obtain the matching index of the selected graduation value in the sequence, and perform weighted operation to obtain the matching index threshold; Step S500: Determine the rationing strategy and the target time period of the target consumables category, obtain the usage records of the medical consumables during the target time period, and conduct a time series evaluation on the centralized characteristics and the dispersed characteristics of the specifications of the medical consumables; Step S600: Perform a weighted operation on the characteristic values corresponding to the centralized characteristics and the dispersed characteristics of the specifications to obtain the consumption characteristic evaluation value. When the difference from the matching index threshold exceeds the difference threshold, send an alarm prompt to the management personnel.

2. The method for tracing the pharmaceutical supply chain based on association rule analysis according to claim 1, wherein: Step S100 includes: Step S101: Take a certain type of medical consumables as the target consumables category, and obtain the usage records of all specifications of the target consumables within the reference time range; Step S102: Divide the reference time range into several unit reference cycles with a time length of T1, and obtain all the usage records of the k-th specification consumables in the target category in the c-th unit reference cycle in the reference time range.

3. The method for tracing the pharmaceutical supply chain based on association rule analysis according to claim 2, characterized in that: Step S200 includes: Step S201: In the c-th unit reference period, obtain all usage times Q of the medical consumables corresponding to the target consumable category v,c and the usage times Q of the k-th specification consumables k,c , divide the c-th unit reference period into i interval segments, and calculate the proportion of the usage times of the k-th specification consumables in each interval segment to all usage times of the medical consumables corresponding to the target consumable category; Step S202: Aggregate the data values of the ratios of all i interval segments, and denote the variance of the data values of the ratios as the fluctuation coefficient D k,c ; Step S203: Calculate the specification weight W of the k-th specification consumable in the c-th unit reference period k,c , ; Step S204: Calculate the usage frequency F of the consumables of the kth specification in the cth unit reference period k,c , F k,c =Q k,c / tc, where tc represents the time length of the cth unit reference period; Step S205: Calculate the centralized reference value A of the medical consumables corresponding to the target consumable category in the c-th unit reference period c , , where n represents the total number of specifications of the medical consumables in the target category.

4. The method for tracing the pharmaceutical supply chain based on association rule analysis according to claim 3, wherein: Step S300 includes: Step S301: Obtain the usage scenarios of the medical consumables corresponding to the target consumables category from the usage records of the target consumables category. The usage scenarios include intensive care, emergency, general outpatient clinic, and chronic disease outpatient clinic, and set the weight factors for each usage scenario; Step S302: Obtain the application scenarios of each use of the k-th specification consumable in the c-th unit reference period, and use the average value of the weight factors as the priority coefficient P of the k-th specification consumable in the c-th unit reference period k,c ; Step S303: In the c-th unit reference period, obtain the number of times N that the consumable of the k-th specification is first selected k,c , the total number of times N that the medical consumables corresponding to the target consumable category are selected v,c , calculate the demand distribution factor R in the c-th unit reference period k,c , ; Step S304: In the c-th unit reference period, obtain the average usage time interval L of the consumables of the k-th specification, and calculate the dispersion factor V of the consumables of the k-th specification in the c-th unit reference period. k,c , where R k,c , , j,c represents the demand distribution factor of the consumables of the j-th specification calculated in the c-th unit reference period; Step S305: Calculate the usage frequency E of the k-th specification consumables in the c-th unit reference period k,c , E k,c = N k,c / tc; Step S306: Calculate the reference value D of the specification dispersion of the medical consumables corresponding to the target consumable category in the c-th unit reference period c , .

5. The pharmaceutical supply chain traceability method based on association rule analysis according to claim 4, characterized in that: Step S400 includes: Step S401: Calculate the matching index M for the c-th unit reference period c , M c = ω1 × A c + ω2 × D c , where ω1 and ω2 represent weights, satisfying the conditions ω1 > 0, ω2 > 0 and ω1 + ω2 = 1; Step S402: Obtain the matching indexes of M unit cycles, and arrange all M matching indexes in ascending order according to the values to obtain the matching index sequence; Step S403: Set the balance coefficient α, satisfying the condition 0.25 ≤ α ≤ 0.75, and calculate the position serial number x in the sequence, x = α × (M + 1); Step S404: When x is an integer, take the x-th matching index in the matching index sequence as the matching index threshold Mt. When x is not an integer, round up and round down x to obtain two integers h and h + 1; Step S405: Use the integer with the smallest absolute value of the difference from x among h and h + 1 as the reference value h0, and calculate the matching index threshold Mt, where Mt = M h0 + α × (M h+1 - M h ).

6. The pharmaceutical supply chain traceability method based on association rule analysis according to claim 5, characterized in that: The steps for performing the time series evaluation of the centralized characteristics of the specifications in Step S500 include: Step S501: Obtain the rationing strategy of the target consumables category during the target time period, take the rationing strategy as the target rationing strategy, and take a time period with the same rationing strategy as the target consumables category as the target time period; Step S502: Divide the target time period into a number of target unit cycles with a time length of T2, where T2 < T1, and obtain all the usage records of the k-th specification consumables in the target category during the d-th target unit cycle of the target time period; Step S503: Calculate the specification weight W of the k-th specification consumable in the target consumable category in the d-th target unit period k,d , obtain the usage frequency F of the k-th specification consumable in the d-th target unit period k,d , calculate the centralized reference value B of the specifications of the medical consumables corresponding to the target consumable category in the d-th target unit period d , ; Step S504: Calculate the temporal feature value S in the (d + 1)-th specification set d+1 , , where λ represents a coefficient satisfying the condition 0 < λ < 1.

7. The method for tracing the origin of the pharmaceutical supply chain based on association rule analysis according to claim 6, wherein: The steps for performing temporal evaluation on the specification dispersion characteristics in step S500 include: Step S505: Obtain the application scenarios of each use of the consumables of the k-th specification in the d-th target unit period, and calculate the demand distribution factor R of the c-th unit reference period k,d ; Step S506: Obtain the average usage time interval L of the consumables of the k-th specification in the d-th target unit period k,d , calculate the dispersion factor V of the consumables of the k-th specification in the d-th target unit period k,d , obtain the usage frequency E of the consumables of the k-th specification in the d-th target unit period k,d ; Step S507 calculates the dispersion reference value G of the medical consumables corresponding to the target consumable category in the d-th target unit period d , ; Step S508 calculates the time series eigenvalue U of the (d + 1)-th specification dispersion feature d+1 , .

8. The pharmaceutical supply chain traceability method based on association rule analysis according to claim 7, characterized in that: Step S600 includes: Step S601: Calculate the (d + 1)-th consumption feature evaluation value γ d+1 , γ d+1 = ω1 × S d+1 + ω2 × U d+1 , Step S602: Set the difference threshold β. When |Mt - γ d+1 | > β, an alarm prompt is sent to the manager of the target consumable category.

9. A pharmaceutical supply chain traceability system based on association rule analysis, which is used to execute the pharmaceutical supply chain traceability method based on association rule analysis according to any one of claims 1-8, and is characterized in that: The system includes: A historical record management module, a historical feature management module, a feature comparison module, and a difference evaluation module; The historical record management module is used to store the usage records of medical consumables. The historical feature management module is used to collect the specification concentration reference value and the specification dispersion reference value in the usage records of medical consumables, and calculate the matching index threshold. The feature comparison module is used to compare the consumption characteristics of medical consumables in the target time period and the reference time range. The difference evaluation module is used to evaluate the degree of difference.

10. The pharmaceutical supply chain traceability system based on association rule analysis according to claim 9, characterized in that: The historical feature management module includes a specification weight management unit, a concentration reference value calculation unit, a demand distribution factor management unit, a dispersion factor management unit, a dispersion reference value calculation unit, a matching index management unit, and a matching index threshold management unit; Among them, the specification weight management unit is used to calculate the specification weight of a certain specification of medical consumables. The concentration reference value calculation unit is used to calculate the specification concentration reference value of medical consumables. The demand distribution factor management unit is used to calculate the demand distribution factor of a certain specification of medical consumables. The dispersion factor management unit is used to calculate the dispersion factor of a certain specification of medical consumables. The dispersion reference value calculation unit is used to calculate the specification dispersion reference value of medical consumables. The matching index management unit is used to calculate the matching index of a unit cycle. The matching index threshold management unit is used to calculate the matching index threshold; The feature comparison module includes a feature value management unit and a temporal feature management unit; Among them, the feature value management unit is used to obtain the specification weight, the specification concentration reference value, the demand distribution factor, the dispersion factor, and the specification dispersion reference value of the target consumable category in the target time period. The temporal feature management unit is used to calculate the temporal feature value of the specification concentration and the temporal feature value of the specification dispersion; The difference evaluation module includes: a consumption feature management unit and an information reminder unit; Among them, the consumption feature management unit is used to calculate the consumption feature evaluation value. The information reminder unit is used to manage the difference threshold and give an information prompt for the determination result that meets the alarm condition.

Images