Large-scale supermarket visitor flow early warning management method and system
By establishing a traffic warning model and monitoring the intensity of staff activities, the problem that large supermarkets cannot obtain traffic data in real time is solved, accurate warning and optimized management of traffic is achieved, operational risks are reduced, and the effectiveness of marketing strategies is improved. .
Patent Information
- Application Number
- CN202411125929.7
- Authority / Receiving Office
- CN · China
- Patent Type
- Applications(China)
- Current Assignee / Owner
- Filing Date
- 2024-08-16
- Publication Date
- 2024-11-22
- Estimated Expiration
- Not applicable · inactive patent
AI Technical Summary
Large supermarkets find it difficult to obtain traffic data in real time, resulting in the inability to allocate staff in time and the inability to effectively predict traffic trends, which may cause safety hazards or operational risks.
By collecting historical and real-time traffic data, establishing a traffic warning model, using cameras to monitor the intensity of staff activities and information on the gathering area of people, formulating standards and optimizing warning models, analyzing extreme traffic situations, and setting up management methods to deal with fluctuations in traffic.
It has achieved accurate warnings on traffic of large supermarkets, optimized human resources allocation, reduced operational risks caused by excessive or too small traffic, and improved the accuracy of marketing strategies and customer attractiveness.
Smart Images

Figure CN119006052A_ABST
Abstract
Description
Technical Field
[0001] The present invention relates to the field of computer technology and management technology, and in particular to a method and system for early warning management of passenger flow in a large supermarket. Background Art
[0002] In real life, the role of crowd warning management in large supermarkets is multifaceted. It can not only enhance customer experience and ensure safety and health, but also optimize operational management, improve sales efficiency, promote market research and enhance brand image.
[0003] Based on the current situation, it is necessary to propose a method and system for early warning management of passenger flow in large supermarkets, which can obtain the current passenger flow in the supermarket in real time, allocate corresponding staff in time, and effectively predict the passenger flow trend to avoid safety hazards caused by excessive passenger flow in large supermarkets and the disadvantages of not being able to discover the problem in time due to too little passenger flow; at the same time, it is also convenient for supermarket management to know the distribution of passenger flow in supermarkets in various time periods, and arrange corresponding staff in advance according to the rules to ensure the normal and orderly operation of the supermarket. Summary of the invention
[0004] In view of the deficiencies in the prior art, the present invention provides a method and system for early warning management of passenger flow in large supermarkets, which solves the problem that large supermarkets cannot obtain the current passenger flow in the supermarket in real time, timely assign corresponding staff, and effectively predict the passenger flow trend, so as to avoid safety hazards caused by excessive passenger flow in large supermarkets and the problem that disadvantages cannot be discovered in time due to insufficient passenger flow.
[0005] To achieve the above objectives, the present invention is implemented through the following technical solutions: A large supermarket traffic flow early warning management method, comprising the following steps:
[0006] S1. Collect data on the flow of people over the years from the local database of large supermarkets as initial samples; use equipment to monitor the real-time flow of people in supermarkets every day, including real-time entrance flow and real-time exit flow, as existing samples; establish an initial warning model for the flow of people in large supermarkets by processing the initial samples and existing samples;
[0007] S2. Set up two groups of cameras, D and E, in the supermarket to collect information about the activity intensity of the staff and the areas where people gather in the supermarket; formulate standards by analyzing the activity intensity of the staff; and count the promotion rate of discount behavior in areas with different density of people gathering on the flow of people in the supermarket;
[0008] S3. Analyze the relationship between the activity intensity of staff and the areas where people gather in supermarkets, as well as the impact on the flow of people in supermarkets, to assist in the optimization of the initial warning model of the flow of people in large supermarkets, so as to generate a dynamic warning model of the flow of people in large supermarkets;
[0009] S4. Based on the crowd flow warning model of large supermarkets, analyze the extreme situations of crowd flow, obtain the main and secondary reasons for the drastic fluctuations in crowd flow, and set up management methods based on the main and secondary reasons.
[0010] As a further solution of the present invention: the step S1, processing the initial sample and the existing sample, comprises the following steps:
[0011] Process the initial sample data, plot the passenger flow data of each year into a curve graph bi, and record the part of the curve graph bi that is higher than the K value as the marked segment kij;
[0012] In the comparison results, if the number of overlapping marked segments in the same time period is greater than or equal to M, the time period is marked as a high warning period;
[0013] In the comparison results, if the number of overlapping marked segments in the same time period is greater than or equal to m and less than M, the time period is marked as a moderate warning period;
[0014] In the comparison results, if the number of overlapping marked segments in the same time period is less than m, the time period is marked as a normal period;
[0015] Get the warning period gradient;
[0016] The initial sample is processed by subtracting the real-time outlet flow from the real-time inlet flow to obtain A;
[0017] Through the warning period gradient and real-time retained flow A, an initial warning model for passenger flow in large supermarkets is established.
[0018] As a further solution of the present invention: Step S1, establishing an initial warning model for the flow of people in a large supermarket, comprises the following steps:
[0019] Combine the processed initial samples and existing samples to calculate the median of the retained flow A obtained daily, draw a curve graph c, and according to the curve graph bi, mark the high warning period, medium warning period and normal period in the curve graph c respectively;
[0020] Set the red line k for human flow, and in the curve graph c, record the time period below the value k as the low tide period; the curve graph c is the initial warning system and model for human flow in large supermarkets.
[0021] As a further solution of the present invention: the step S2, analyzing the activity intensity of the staff, thereby formulating the standard, comprises the following steps:
[0022] According to the flow of people during the normal period, three classification standards w1, w2 and w3 are set up. That is, under this standard, by counting the displacement frequency P and displacement distance S of the staff, the average work intensity W of the staff under each standard is estimated, and the expression is:
[0023] W = P*S*E;
[0024] Where E is the energy consumption of the worker per meter of displacement, which is an estimated value;
[0025] Under the w3 standard, the displacement frequency P and displacement distance S of each staff member during the high warning period are collected, and the daily work intensity is calculated through the work intensity expression, and the median is taken as the standard work intensity under the w3 standard;
[0026] Under the w2 standard, the displacement frequency P and displacement distance S of each staff member during the moderate warning period are collected, and the daily work intensity is calculated through the work intensity expression, and the median is taken as the standard work intensity under the w2 standard;
[0027] Under the w1 standard, the displacement frequency P and displacement distance S of each staff member during the normal period and the low tide period are collected, and the daily work intensity is calculated through the work intensity expression. The median is taken as the standard work intensity under the w1 standard.
[0028] As a further solution of the present invention: the step S2, which counts the promotion rate of discount behavior on the flow of people in the supermarket in areas with different density of people flow, includes the following steps:
[0029] The cameras in group E are used to monitor the gathering areas of human traffic. The areas where the density of human traffic is higher than the F value for a long time are recorded as high-frequency gathering areas; the rest of the areas are recorded as normal areas;
[0030] The types of goods in the high-frequency gathering area and the types of goods in the normal area are counted respectively. In the traffic warning models of large supermarkets in the high-frequency gathering area and the normal area, the changes in the traffic flow in the supermarkets during the time period of goods discounts are calculated. The changes in the traffic flow in the non-marked section are used as the main reference data, with a weight of Q%; the changes in the traffic flow in the marked section are used as auxiliary reference data, with a weight of q%; the promotion rate G of the discount in the high-frequency gathering area on the traffic flow in the supermarket and the promotion rate H of the discount in the normal area on the traffic flow in the supermarket are calculated;
[0031] Calculate the average daily traffic volume r of supermarkets in the normal area, assuming that the discount period is t days, the traffic volume in the high-frequency gathering area within t days is g, the discount intensity coefficient is α, and the traffic volume in the low-frequency gathering area within t days is h, and the discount intensity coefficient is β;
[0032] G=[(gr*t) / (r*t)]*α*100%;
[0033] H=[(hr*t) / (r*t)]*β*100%;
[0034] As a further solution of the present invention: Step S3, establishing a dynamic large-scale supermarket traffic warning model, includes the following steps:
[0035] Based on the statistical data of the daily work intensity of the staff, the areas are divided according to three classification standards:
[0036] The area where the work intensity is greater than the w3 standard is divided into the adjustment area;
[0037] The area where the work intensity is greater than the w2 standard and less than or equal to the w3 standard is divided into the observation area;
[0038] The area with work intensity less than or equal to w2 standard and greater than w1 standard is divided into ordinary area;
[0039] The area with work intensity less than or equal to w1 standard is divided into red line area;
[0040] Mark the adjustment area in the curve graph c, compare it with the original graph, divide the non-overlapping area between the adjustment area and the high warning period, search for discount information in the non-overlapping area, and if G is less than G' and H is less than H', divide the area into the high warning period;
[0041] Mark the observation area in the curve graph c, compare it with the original graph, divide the overlap area between the observation area and the high warning period, search for discount information in the overlap area, and if G is greater than or equal to G' and H is greater than or equal to H', then divide the area into the moderate warning period;
[0042] Mark the normal area in the curve graph c, compare it with the original graph, divide the overlapped area between the normal area and the low tide area, search for discount information in the overlapped area, and if G is greater than or equal to G' and H is greater than or equal to H', then divide the area into the low tide period;
[0043] On the curve graph c, the high warning period, medium warning period, normal period and low tide period are re-divided to establish a dynamic pedestrian flow warning model for large supermarkets.
[0044] As a further solution of the present invention: Step S4, establishing a dynamic large-scale supermarket traffic warning model, includes the following steps:
[0045] When the flow of people in supermarkets tends to be too high, if the main reason is that the discount in the supermarket is too high, and the secondary reason is that it is in the high warning period or medium warning period, contact the relevant merchants to adjust the discount appropriately;
[0046] If the main reason is that the store is in the high warning period or the medium warning period, and the secondary reason is that the discount in the supermarket is too high, then increase the number of staff in the supermarket, offer limited-time discounts, and encourage different types of customers to visit in batches;
[0047] When the flow of people in a supermarket is trending too low, if the main reason is that the discount in the supermarket is too low, and the secondary reason is that it is in a low tide period, contact the relevant merchants to increase the discount appropriately;
[0048] If the main reason is that the business is in a low tide period and the secondary reason is that the discount in the supermarket is too low, then optimize the layout and decoration of the supermarket, focus on supermarket marketing for merchants in high-frequency gathering areas, and provide relevant training for staff.
[0049] As another aspect of the present invention: a large supermarket human flow early warning management system, which includes: the system specifically adopts a large supermarket human flow early warning management method to realize early warning management of human flow.
[0050] Beneficial Effects
[0051] The present invention provides a method for early warning management of passenger flow in large supermarkets. Compared with the prior art, it has the following beneficial effects:
[0052] The present invention establishes a crowd flow warning model by collecting and processing historical and real-time crowd flow data; collects information on staff activity intensity and crowd flow gathering areas through cameras, and formulates corresponding standards and promotion rates; counts staff work intensity and discount area promotion rate to crowd flow, and further improves the crowd flow warning model; analyzes the causes of extreme crowd flow situations based on the warning model, and sets management methods;
[0053] It optimizes the allocation of human resources, provides accurate marketing strategies for supermarkets, increases customer appeal and sales, predicts and responds to drastic fluctuations in traffic, and reduces operational risks caused by excessive or insufficient traffic. BRIEF DESCRIPTION OF THE DRAWINGS
[0054] Figure 1 is a flow chart of the method of the present invention;
[0055] Figure 2 It is a system block diagram of the present invention. DETAILED DESCRIPTION
[0056] The following will be combined with the drawings in the embodiments of the present invention to clearly and completely describe the technical solutions in the embodiments of the present invention. Obviously, the described embodiments are only part of the embodiments of the present invention, not all of the embodiments. Based on the embodiments of the present invention, all other embodiments obtained by ordinary technicians in this field without creative work are within the scope of protection of the present invention.
[0057] Embodiment 1
[0058] See also Figure 1 , this application provides a large supermarket traffic warning management method, including:
[0059] S1. Collect data on the flow of people over the years from the local database of large supermarkets as initial samples; use equipment to monitor the real-time flow of people in supermarkets every day, including real-time entrance flow and real-time exit flow, as existing samples; establish an initial warning model for the flow of people in large supermarkets by processing the initial samples and existing samples;
[0060] Process the initial sample data, plot the passenger flow data of each year into a curve graph bi, and record the part of the curve graph bi that is higher than the K value as the marked segment kij;
[0061] In the comparison results, if the number of overlapping marked segments in the same time period is greater than or equal to M, the time period is marked as a high warning period;
[0062] In the comparison results, if the number of overlapping marked segments in the same time period is greater than or equal to m and less than M, the time period is marked as a moderate warning period;
[0063] In the comparison results, if the number of overlapping marked segments in the same time period is less than m, the time period is marked as a normal period;
[0064] Get the warning period gradient;
[0065] The initial sample is processed by subtracting the real-time outlet flow from the real-time inlet flow to obtain A;
[0066] Through the warning period gradient and real-time retained flow A, an initial warning model for the flow of people in large supermarkets is established;
[0067] Combine the processed initial samples and existing samples to calculate the median of the retained flow A obtained daily, draw a curve graph c, and according to the curve graph bi, mark the high warning period, medium warning period and normal period in the curve graph c respectively;
[0068] Set a red line k for the flow of people, and in the curve graph c, record the time period below the k value as the low tide period; the curve graph c is the initial warning system and model for the flow of people in large supermarkets;
[0069] The establishment of the early warning period gradient can highlight the key points through the changes in the graph, allowing observers to pay more attention to the time periods that need special attention, grasp the data changes of the flow of people in different early warning periods, and enable observers to quickly understand the overall trend and characteristics of the flow of people.
[0070] The values of K, k, M and m are set by the relevant staff of the supermarket based on experience.
[0071] S2. Set up two groups of cameras, D and E, in the supermarket to collect information about the activity intensity of the staff and the areas where people gather in the supermarket; formulate standards by analyzing the activity intensity of the staff; and count the promotion rate of discount behavior in areas with different density of people gathering on the flow of people in the supermarket;
[0072] According to the flow of people during the normal period, three classification standards w1, w2 and w3 are set up. That is, under this standard, by counting the displacement frequency P and displacement distance S of the staff, the average work intensity W of the staff under each standard is estimated, and the expression is:
[0073] W = P*S*E;
[0074] Where E is the energy consumption of the worker per meter of displacement, which is an estimated value;
[0075] Under the w3 standard, the displacement frequency P and displacement distance S of each staff member during the high warning period are collected, and the daily work intensity is calculated through the work intensity expression, and the median is taken as the standard work intensity under the w3 standard;
[0076] Under the w2 standard, the displacement frequency P and displacement distance S of each staff member during the moderate warning period are collected, and the daily work intensity is calculated through the work intensity expression, and the median is taken as the standard work intensity under the w2 standard;
[0077] Under the w1 standard, the displacement frequency P and displacement distance S of each worker during the normal period and the low tide period are collected, and the daily work intensity is calculated through the work intensity expression, and the median is taken as the standard work intensity under the w1 standard;
[0078] The cameras in group E are used to monitor the gathering areas of human traffic. The areas where the density of human traffic is higher than the F value for a long time are recorded as high-frequency gathering areas; the rest of the areas are recorded as normal areas;
[0079] The types of goods in the high-frequency gathering area and the types of goods in the normal area are counted respectively. In the traffic warning models of large supermarkets in the high-frequency gathering area and the normal area, the changes in the traffic flow in the supermarkets during the time period of goods discounts are calculated. The changes in the traffic flow in the non-marked section are used as the main reference data, with a weight of Q%; the changes in the traffic flow in the marked section are used as auxiliary reference data, with a weight of q%; the promotion rate G of the discount in the high-frequency gathering area on the traffic flow in the supermarket and the promotion rate H of the discount in the normal area on the traffic flow in the supermarket are calculated;
[0080] Calculate the average daily traffic volume r of supermarkets in the normal area, assuming that the discount period is t days, the traffic volume in the high-frequency gathering area within t days is g, the discount intensity coefficient is α, and the traffic volume in the low-frequency gathering area within t days is h, and the discount intensity coefficient is β;
[0081] G=[(gr*t) / (r*t)]*α*100%;
[0082] H=[(hr*t) / (r*t)]*β*100%.
[0083] S3. Analyze the relationship between the activity intensity of staff and the areas where people gather in supermarkets, as well as the impact on the flow of people in supermarkets, to assist in the optimization of the initial warning model of the flow of people in large supermarkets, so as to generate a dynamic warning model of the flow of people in large supermarkets;
[0084] Based on the statistical data of the daily work intensity of the staff, the areas are divided according to three classification standards:
[0085] The area where the work intensity is greater than the w3 standard is divided into the adjustment area;
[0086] The area where the work intensity is greater than the w2 standard and less than or equal to the w3 standard is divided into the observation area;
[0087] The area with work intensity less than or equal to w2 standard and greater than w1 standard is divided into ordinary area;
[0088] The area with work intensity less than or equal to w1 standard is divided into red line area;
[0089] Mark the adjustment area in the curve graph c, compare it with the original graph, divide the non-overlapping area between the adjustment area and the high warning period, search for discount information in the non-overlapping area, and if G is less than G' and H is less than H', divide the area into the high warning period;
[0090] Mark the observation area in the curve graph c, compare it with the original graph, divide the overlap area between the observation area and the high warning period, search for discount information in the overlap area, and if G is greater than or equal to G' and H is greater than or equal to H', then divide the area into the moderate warning period;
[0091] Mark the normal area in the curve graph c, compare it with the original graph, divide the overlapped area between the normal area and the low tide area, search for discount information in the overlapped area, and if G is greater than or equal to G' and H is greater than or equal to H', then divide the area into the low tide period;
[0092] The high warning period, medium warning period, normal period and low tide period on the curve c are re-divided to establish a dynamic large supermarket traffic warning model;
[0093] Among them, G' and H' are set by the operator based on experience;
[0094] By comprehensively considering the work intensity of employees in the mall and the retention intention of customers, the gradient division of the warning period is optimized and adjusted. Such corrections will make the prediction standard more objective and the prediction results more accurately reflect the changes in real-time traffic flow. It also improves the prediction accuracy of the dynamic traffic warning model for large supermarkets, making it more in line with the current actual situation of supermarkets.
[0095] S4. Based on the crowd flow warning model of large supermarkets, analyze the extreme situations of crowd flow, obtain the main and secondary reasons for the drastic fluctuations in crowd flow, and set up management methods based on the main and secondary reasons.
[0096] When the flow of people in supermarkets tends to be too high, if the main reason is that the discount in the supermarket is too high, and the secondary reason is that it is in the high warning period or medium warning period, contact the relevant merchants to adjust the discount appropriately;
[0097] If the main reason is that the store is in the high warning period or the medium warning period, and the secondary reason is that the discount in the supermarket is too high, then increase the number of staff in the supermarket, offer limited-time discounts, and encourage different types of customers to visit in batches;
[0098] When the flow of people in a supermarket is trending too low, if the main reason is that the discount in the supermarket is too low, and the secondary reason is that it is in a low tide period, contact the relevant merchants to increase the discount appropriately;
[0099] If the main reason is that the business is in a low tide period and the secondary reason is that the discount in the supermarket is too low, then optimize the layout and decoration of the supermarket, and focus on supermarket marketing for merchants in high-frequency gathering areas; provide relevant training for staff, and provide efficient and friendly customer service, including fast checkout channels, free packaging services, etc., to improve customer satisfaction.
[0100] Embodiment 2
[0101] See also Figure 2 A large supermarket passenger flow early warning management system, the system uses the above-mentioned large supermarket passenger flow early warning management method to perform early warning management on the large supermarket passenger flow, the system comprises:
[0102] The data collection unit is used to search for relevant data in the local database of the supermarket and mark it as the initial sample; use the infrared two-way sensor to obtain the real-time flow of people in the supermarket every day and mark it as the existing sample; and transmit the initial sample and the existing sample to the statistical unit in the form of electrical signals;
[0103] The statistical comparison unit is used to process the initial sample and generate a curve graph bi. With the help of the correction signal, the curve graph c is generated from bi and the warning period is divided therein; and the curve graph c is transmitted to the storage unit and the display unit in the form of an electrical signal;
[0104] A storage unit, used for receiving and storing data transmitted from the statistical comparison unit and the real-time image capture unit, and receiving a data call signal from the feedback unit;
[0105] The real-time image capture unit is used to capture the displacement data of the staff inside the supermarket, so as to formulate and update the daily work intensity standard of the staff; record the flow of people and gathering areas in the supermarket, and update the promotion rate of the supermarket flow when the gathering areas are discounted every day; finally, the above data is transmitted to the feedback unit and the display unit in the form of electrical signals;
[0106] A feedback unit, used for calling the data of the storage unit to perform standard correction on the data from the real-time image capture unit, generating a correction signal and transmitting it to the statistical comparison unit and the real-time image capture unit;
[0107] Management unit, used to analyze the flow trend of people in supermarkets and provide management suggestions through big data;
[0108] The display unit is used to display the curve graph c of the statistical comparison unit and the flow trend analysis and management suggestions.
[0109] Some of the data in the above formulas are dimensionless and numerically calculated. Meanwhile, the contents not described in detail in this specification belong to the prior art known to those skilled in the art.
[0110] The above embodiments are only used to illustrate the technical method of the present invention rather than to limit it. Although the present invention has been described in detail with reference to the preferred embodiments, those skilled in the art should understand that the technical method of the present invention may be modified or replaced by equivalents without departing from the spirit and scope of the technical method of the present invention.
Claims
1. A method for early warning management of passenger flow in large supermarkets, characterized in that: The following steps are involved: S1. Collect data on the flow of people over the years from the local database of large supermarkets as the initial sample; use equipment to monitor the real-time flow of people in the supermarket every day, including real-time entrance flow and real-time exit flow, as the existing sample; By processing the initial samples and existing samples, an initial warning model for the flow of people in large supermarkets is established; S2. Set up two groups of cameras, D and E, in the supermarket to collect information about the activity intensity of the staff and the areas where people gather in the supermarket; formulate standards by analyzing the activity intensity of the staff; and count the promotion rate of discount behavior in areas with different densities of people gathering on the flow of people in the supermarket; S3. Analyze the relationship between the activity intensity of staff and the areas where people gather in supermarkets, as well as the impact on the flow of people in supermarkets, to assist in the optimization of the initial warning model of the flow of people in large supermarkets, so as to generate a dynamic warning model of the flow of people in large supermarkets; S4. Based on the crowd flow warning model of large supermarkets, analyze the extreme situations of crowd flow, obtain the main and secondary reasons for the drastic fluctuations in crowd flow, and set up management methods based on the main and secondary reasons.
2. A large supermarket passenger flow early warning management method according to claim 1, characterized in that: By processing the initial samples and existing samples, an initial warning model for the flow of people in large supermarkets is established, including the following steps: Process the initial sample data, plot the passenger flow data of each year into a curve graph bi, and record the part of the curve graph bi that is higher than the K value as the marked segment kij; In the comparison results, if the number of overlapping marked segments in the same time period is greater than or equal to M, the time period is marked as a high warning period; In the comparison results, if the number of overlapping marked segments in the same time period is greater than or equal to m and less than M, the time period is marked as a moderate warning period; In the comparison results, if the number of overlapping marked segments in the same time period is less than m, the time period is marked as a normal period; Get the warning period gradient; The initial sample is processed by subtracting the real-time outlet flow from the real-time inlet flow to obtain A; Through the warning period gradient and real-time retained flow A, an initial warning model for passenger flow in large supermarkets is established.
3. A large supermarket traffic early warning management method according to claim 2, characterized in that: By combining the warning period gradient and analyzing the retained traffic A, an initial warning model for the flow of people in large supermarkets is established, including the following steps: Combine the processed initial samples and existing samples to calculate the median of the retained flow A obtained daily, draw a curve graph c, and according to the curve graph bi, mark the high warning period, medium warning period and normal period in the curve graph c respectively; Set the red line k for human flow, and in the curve graph c, record the time period below the value k as the low tide period; the curve graph c is the initial warning system and model for human flow in large supermarkets.
4. A large supermarket passenger flow early warning management method according to claim 1, characterized in that: By analyzing the activity intensity of the staff, the standards are established, including the following steps: According to the flow of people during the normal period, three classification standards w1, w2 and w3 are set up. That is, under this standard, by counting the displacement frequency P and displacement distance S of the staff, the average work intensity W of the staff under each standard is estimated, and the expression is: W = P*S*E; Where E is the energy consumption of the worker per meter of displacement, which is an estimated value; Under the w3 standard, the displacement frequency P and displacement distance S of each staff member during the high warning period are collected, and the daily work intensity is calculated through the work intensity expression, and the median is taken as the standard work intensity under the w3 standard; Under the w2 standard, the displacement frequency P and displacement distance S of each staff member during the moderate warning period are collected, and the daily work intensity is calculated through the work intensity expression, and the median is taken as the standard work intensity under the w2 standard; Under the w1 standard, the displacement frequency P and displacement distance S of each staff member during the normal period and the low tide period are collected, and the daily work intensity is calculated through the work intensity expression. The median is taken as the standard work intensity under the w1 standard.
5. A large supermarket passenger flow early warning management method according to claim 1, characterized in that: Counting the promotion rate of discount behavior on the flow of people in supermarkets in areas with different density of people flow, including the following steps: The cameras in group E are used to monitor the gathering areas of human traffic. The areas where the density of human traffic is higher than the F value for a long time are recorded as high-frequency gathering areas; the rest of the areas are recorded as normal areas; The types of goods in the high-frequency gathering area and the types of goods in the normal area are counted respectively. In the traffic warning models of large supermarkets in the high-frequency gathering area and the normal area, the changes in the traffic flow in the supermarkets during the time period of goods discounts are calculated. The changes in the traffic flow in the non-marked section are used as the main reference data, with a weight of Q%; the changes in the traffic flow in the marked section are used as auxiliary reference data, with a weight of q%; the promotion rate G of the discount in the high-frequency gathering area on the traffic flow in the supermarket and the promotion rate H of the discount in the normal area on the traffic flow in the supermarket are calculated; Calculate the average daily traffic volume r of supermarkets in the normal area, assuming that the discount period is t days, the traffic volume in the high-frequency gathering area within t days is g, the discount intensity coefficient is α, and the traffic volume in the low-frequency gathering area within t days is h, and the discount intensity coefficient is β; G=[(gr*t) / (r*t)]*α*100%; H=[(hr*t) / (r*t)]*β*100%.
6. A large supermarket passenger flow early warning management method according to claim 1, characterized in that: Establishing a dynamic large-scale supermarket traffic flow warning model includes the following steps: Based on the statistical data of the daily work intensity of the staff, the areas are divided according to three classification standards: The area where the work intensity is greater than the w3 standard is divided into the adjustment area; The area where the work intensity is greater than the w2 standard and less than or equal to the w3 standard is divided into the observation area; The area with work intensity less than or equal to w2 standard and greater than w1 standard is divided into ordinary area; The area with work intensity less than or equal to w1 standard is divided into red line area; Mark the adjustment area in the curve graph c, compare it with the original graph, divide the non-overlapping area between the adjustment area and the high warning period, search for discount information in the non-overlapping area, and if G is less than G' and H is less than H', divide the area into the high warning period; Mark the observation area in the curve graph c, compare it with the original graph, divide the overlap area between the observation area and the high warning period, search for discount information in the overlap area, and if G is greater than or equal to G' and H is greater than or equal to H', then divide the area into the moderate warning period; Mark the normal area in the curve graph c, compare it with the original graph, divide the overlapped area between the normal area and the low tide area, search for discount information in the overlapped area, and if G is greater than or equal to G' and H is greater than or equal to H', then divide the area into the low tide period; On the curve graph c, the high warning period, medium warning period, normal period and low tide period are re-divided to establish a dynamic pedestrian flow warning model for large supermarkets.
7. A large supermarket passenger flow early warning management method according to claim 1, characterized in that: Set up management methods for major and minor causes, including the following steps: When the flow of people in supermarkets tends to be too high, if the main reason is that the discount in the supermarket is too high, and the secondary reason is that it is in the high warning period or medium warning period, contact the relevant merchants to adjust the discount appropriately; If the main reason is that the store is in the high warning period or the medium warning period, and the secondary reason is that the discount in the supermarket is too high, then increase the number of staff in the supermarket, offer limited-time discounts, and encourage different types of customers to visit in batches; When the flow of people in a supermarket is trending too low, if the main reason is that the discount in the supermarket is too low, and the secondary reason is that it is in a low tide period, contact the relevant merchants to increase the discount appropriately; If the main reason is that the business is in a low tide period and the secondary reason is that the discount in the supermarket is too low, then optimize the layout and decoration of the supermarket, focus on supermarket marketing for merchants in high-frequency gathering areas, and provide relevant training for staff.
8. A large supermarket passenger flow warning management system, characterized in that: The system includes: Data collection unit: used to search for relevant data in the local database of the supermarket and mark it as the initial sample; use infrared two-way sensors to obtain the real-time flow of people in the supermarket every day and mark it as the existing sample; and transmit the initial sample and the existing sample to the statistical unit in the form of electrical signals; Statistical comparison unit: used to process the initial sample and make a curve graph bi. With the help of the correction signal, a curve graph c is generated from bi, and the warning period is divided in it, and it is transmitted to the storage unit and the display unit in the form of an electrical signal; Storage unit: used to receive and store data from the statistical comparison unit and the real-time image capture unit, and receive data call signals from the feedback unit; Real-time image capture unit: used to capture the displacement data of the staff inside the supermarket, so as to formulate and update the daily work intensity standard of the staff; record the flow of people and gathering areas in the supermarket, and update the promotion rate of the supermarket flow when the gathering areas are discounted every day; finally, transmit the above data to the feedback unit and display unit in the form of electrical signals; A feedback unit, used for calling the data of the storage unit to perform standard correction on the data from the real-time image capture unit, generating a correction signal and transmitting it to the statistical comparison unit and the real-time image capture unit; Management unit: used to analyze the flow trend of people in supermarkets and provide management suggestions through big data; Display unit; used to display the curve graph c of the statistical comparison unit and the flow trend analysis and management suggestions.
