Intelligent vending machine and remote management system

Abstract

The invention relates to an intelligent vending machine and a remote management system. The intelligent vending machine and the remote management system comprise the intelligent vending machine and a remote management platform. The intelligent vending machine is provided with a movable goods shelf or a mechanical arm used for storing and dynamically adjusting the positions of commodities. The sensor is used for monitoring commodity inventory, mechanical part states and environmental parameters, and is in bidirectional communication with the remote management platform through the communication module. The remote management platform comprises a commodity configuration module, generates a commodity position adjustment instruction based on real-time sales data, sales promotion activities or customer behavior analysis, and sends the instruction to the intelligent vending machine to realize real-time commodity layout adjustment. The system further comprises a fault prevention and self-repair module which analyzes sensor data through machine learning, predicts potential faults of the mechanical part and executes self-diagnosis. And the customer service module provides a video call function and realizes purchase suggestion, fault processing and refunding and changing services. According to the invention, intelligentization and adaptive optimization of commodity management are realized, and the operation efficiency and user experience of the vending machine are improved.

Description

Technical Field

[0001] The present invention relates to the technical field of intelligent vending machines, and in particular to an intelligent vending machine and a remote management system. Background Art

[0002] In the existing technology, smart vending machines usually have the functions of commodity storage, automatic vending and basic inventory management. Some vending machine systems are equipped with sensors to monitor inventory status and environmental conditions, and have basic communication capabilities to connect to remote management platforms. Remote management platforms are usually used to collect sales data, send replenishment instructions, and perform basic remote maintenance operations. In addition, some vending machines have also introduced simple artificial intelligence functions, such as inventory forecasting through historical data or promotion adjustments based on simple rules.

[0003] However, the existing intelligent vending machine system has many limitations, mainly manifested in the lack of flexibility in product management, inaccurate prediction and maintenance of equipment failures, and lack of deep interaction with customers. For example, the arrangement and adjustment of product positions in existing vending machines are usually fixed, making it difficult to flexibly and dynamically optimize according to customer needs. At the same time, the fault diagnosis of mechanical components is mostly passive, and real-time preventive maintenance is impossible.

[0004] Therefore, there is an urgent need to develop a new type of intelligent vending machine and remote management system. Summary of the invention

[0005] The present application provides a smart vending machine and a remote management system to improve the operating efficiency and user experience of the vending machine.

[0006] The present application provides an intelligent vending machine and a remote management system, including:

[0007] An intelligent vending machine, the vending machine comprising a movable shelf or a mechanical arm for storing and rearranging the positions of commodities; sensors for monitoring the inventory status of commodities inside the vending machine, the operating status of mechanical components and environmental parameters; and a communication module for communicating with a remote management platform;

[0008] A remote management platform is configured to communicate bidirectionally with the smart vending machine via a wireless communication network. The remote management platform includes a product configuration module for generating product position adjustment instructions based on real-time sales data, promotional activities or customer behavior analysis. The product position adjustment instructions are sent to the smart vending machine to control the mechanical arm or movable shelf of the smart vending machine so that the vending machine can rearrange the products according to real-time needs; a fault prevention and self-repair module is used to analyze the sensor data in the vending machine using a machine learning algorithm to predict potential failures of mechanical components; when an abnormality is predicted, the vending machine is instructed to perform self-diagnosis and generate a repair request; a customer service module allows customers to communicate with remote customer service personnel in real time through the video call device on the vending machine to provide purchase recommendations, troubleshooting and return and exchange services; the customer service module also includes an intelligent customer service sub-module for responding to common customer questions.

[0009] Furthermore, the intelligent vending machine includes an environment adaptation module, and the environment adaptation module is specifically used for:

[0010] Real-time monitoring of environmental parameters around the vending machine through sensors, including temperature, humidity, light intensity and noise level;

[0011] Dynamically adjust the refrigeration system, lighting system and audio system inside the vending machine based on the monitored environmental parameters;

[0012] When the ambient noise level is detected to be high, the audio output volume and frequency range of the vending machine are adjusted to ensure that the user can clearly hear the voice prompts;

[0013] Adjust the brightness and contrast of the vending machine display according to changes in light intensity to optimize the user's visual experience;

[0014] Based on changes in ambient temperature and humidity, the storage temperature and humidity of different types of goods are intelligently adjusted to maintain optimal quality.

[0015] Furthermore, the remote management platform includes a cross-selling optimization module, and the cross-selling optimization module is used to:

[0016] Analyze sales data and user purchasing behavior at different vending machine locations to identify correlations between products;

[0017] Generate personalized product combination recommendations for each vending machine based on the identified product associations;

[0018] Based on the generated personalized product combination recommendation, the movable shelf or robotic arm of the smart vending machine is guided to place the associated products in adjacent locations.

[0019] Furthermore, the intelligent vending machine includes a dynamic pricing module, and the dynamic pricing module is used to:

[0020] Monitor surrounding environmental factors, including weather conditions, surrounding traffic, and nearby events;

[0021] Dynamically adjust commodity prices based on monitored environmental factors and commodity inventory conditions;

[0022] Launch promotional strategies under certain conditions;

[0023] Synchronize price adjustments and promotion information to vending machine displays and remote management platforms in real time.

[0024] Furthermore, the remote management platform includes a group intelligence collaboration module, and the group intelligence collaboration module is used to:

[0025] Multiple smart vending machines with close geographical locations form a collaborative network;

[0026] Real-time analysis of inventory status and sales trends of each vending machine in the network;

[0027] When it is detected that the stock of a specific commodity in a designated smart vending machine is insufficient but the stock of a nearby smart vending machine is sufficient, a commodity allocation plan is generated;

[0028] According to the commodity allocation plan, unmanned delivery vehicles are called to allocate commodities.

[0029] Furthermore, the intelligent vending machine includes a multimodal interaction module, and the multimodal interaction module is configured as follows:

[0030] Integrate speech recognition, gesture recognition and facial expression analysis technologies;

[0031] Allow users to interact with the vending machine through voice commands, gestures or facial expressions;

[0032] Adjust the vending machine’s response and tone based on the user’s interaction style and emotional state;

[0033] For users with limited mobility, the robot arm can be controlled by voice or gesture to take out the goods;

[0034] Record and analyze users' interaction preferences to personalize and optimize each user's shopping experience.

[0035] Furthermore, the remote management platform includes a social marketing and user participation module, and the social marketing and user participation module is configured as follows:

[0036] Generate a unique social media account for each smart vending machine;

[0037] Automatically publish information such as smart vending machine locations, inventory updates, promotions, etc. to social media platforms;

[0038] Allow users to interact with the vending machine through social media platforms, including pre-ordering items or participating in online voting to decide the next batch of stock;

[0039] Analyze user preferences and adjust product configuration based on social media interaction data;

[0040] Implement location-based social gamification marketing strategies, including punching in to unlock special offers or discounts.

[0041] Furthermore, the smart vending machine includes a product expiration monitoring module, which periodically scans the expiration information of the products in the vending machine based on internal sensors and product label information; when it is detected that the product is about to expire, an alarm is sent to the management terminal through the remote management platform, and discount information or promotional activities are automatically generated according to the expiration date of the product to speed up the sales of the products that are about to expire.

[0042] Furthermore, the remote management platform includes an adaptive commodity layout and robotic arm control module, which realizes the optimal layout of commodities inside the vending machine through the following formula 1:

[0043]

[0044] Among them, L opt is the optimal layout solution set of goods inside the vending machine; m is the total number of goods in the vending machine; W i is the weight coefficient of commodity i, reflecting the impact of the commodity weight on the operation of the robot arm; B i is the average daily sales volume of commodity i; B max is the maximum average daily sales volume of the goods in the vending machine; G i is the customer's interaction frequency with product i, including touch, gaze, and voice command; G th is the threshold of interaction frequency; H i is the current shelf height position of product i; H max is the maximum shelf height in the vending machine; α,μ,v,ρ i is the adjustment coefficient;

[0045] After calculating the optimal layout solution set L opt After that, according to L opt Generate a rearrangement instruction for the goods in the vending machine; and control the mechanical arm to adjust the position of the goods according to the rearrangement instruction.

[0046] Furthermore, the remote management platform includes an integrated inventory and pricing optimization module, which simultaneously calculates the optimal inventory configuration set S of the goods through the following formula 2: opt and the optimal price set P opt :

[0047]

[0048] Among them, S opt is the optimal inventory configuration set, which represents the optimal inventory configuration set of each commodity in the vending machine, S opt ={S1,S2,…S i …,S n}, where S i represents the optimal inventory configuration of the i-th product in the vending machine; P opt is the optimal commodity price set, P opt = {P1,P2,…P i …,P n}, where S i represents the optimal price of the i-th product in the vending machine; n is the total number of products in the vending machine; C s is the commodity shortage cost coefficient; D i is the historical demand for commodity i; I i is the current inventory of commodity i; C h I is the commodity holding cost coefficient; i is the current inventory of product i; max is the maximum inventory capacity of commodity i; E i is the number of days until product i expires; E th The threshold number of days to process expired products in advance; D i is the historical demand for commodity i; D max is the maximum demand for commodity i; w1, w2, λ, λ1, κ, κ1 are adjustment coefficients;

[0049] For the i-th product in the vending machine, according to the calculated optimal inventory configuration S i , generate a replenishment plan and carry out replenishment;

[0050] According to the calculated optimal price P i , adjust the price of goods in the vending machine and update it in real time on the vending machine display.

[0051] The technical solution provided by this application has the following beneficial effects:

[0052] (1) Movable shelves or robotic arms can dynamically adjust the position of goods, realizing intelligent product layout based on real-time sales data, promotional activities or customer behavior. This flexible product management method not only improves the accessibility of goods, but also effectively optimizes product display and improves sales conversion rate. (2) Through the fault prevention and self-repair module, the system uses machine learning algorithms to analyze sensor data in real time. The system can predict potential failures of mechanical components and automatically trigger self-diagnosis and repair requests before abnormalities occur. Compared with traditional passive maintenance methods, this system significantly reduces the downtime of vending machines and improves overall operating efficiency. (3) The customer service module realizes real-time communication between customers and remote customer service personnel through video calls and intelligent customer service functions. This real-time interactive function can not only provide purchase recommendations and troubleshooting, but also handle return and exchange services, providing customers with a more personalized and efficient service experience. (4) Through the two-way communication function of the remote management platform, the system can monitor the inventory, operating status and environmental parameters of the vending machine in real time in the background, and send product position adjustment and maintenance instructions based on the analysis results, simplifying the remote management process and reducing manual maintenance costs. BRIEF DESCRIPTION OF THE DRAWINGS

[0053] Figure 1 It is a schematic diagram of an intelligent vending machine and a remote management system provided in the first embodiment of the present application. DETAILED DESCRIPTION

[0054] Many specific details are described in the following description to facilitate a full understanding of the present application. However, the present application can be implemented in many other ways than those described herein, and those skilled in the art can make similar generalizations without violating the connotation of the present application, so the present application is not limited by the specific implementation disclosed below.

[0055] The first embodiment of the present application provides a smart vending machine and a remote management system. Figure 1 , which is a schematic diagram of the first embodiment of the present application. Figure 1 The first embodiment of the present application provides a smart vending machine and a remote management system for detailed description.

[0056] The smart vending machine and remote management system includes a smart vending machine 101 and a remote management platform 102 .

[0057] The smart vending machine 101 includes a movable shelf or a mechanical arm for storing and rearranging the positions of goods; sensors for monitoring the inventory status of goods inside the vending machine, the operating status of mechanical components and environmental parameters; and a communication module for communicating with a remote management platform.

[0058] The intelligent vending machine 101 is the core device of the system, which is used to realize the storage, distribution, monitoring and adaptive management of commodities. The vending machine has multiple functional components and sub-modules to ensure its efficient, intelligent and flexible operation.

[0059] The key components of the smart vending machine 101 include movable shelves or robotic arms. These components are used to store, sort and rearrange the goods inside the vending machine. The movable shelves are composed of a multi-layer structure, each layer of which has the ability to move horizontally or vertically, and the position is adjusted through a built-in motor or transmission device. The robotic arm is a multi-axis linkage mechanism, usually with at least three degrees of freedom, and can move in the three-dimensional space inside the vending machine. The robotic arm is equipped with a gripping device that can accurately grasp, carry and rearrange goods. The control system of the shelf or robotic arm automatically adjusts its motion trajectory based on the received product position adjustment instructions so that the goods are placed in the optimal position, thereby optimizing the customer's purchasing experience and promoting the efficient sales of goods.

[0060] The smart vending machine 101 is also equipped with a variety of sensors to achieve real-time monitoring of the internal environment and equipment operating status of the vending machine. Inventory status sensors include weight sensors, photoelectric sensors and barcode scanners, which are used to detect the inventory and location of various commodities. The operating status of mechanical parts is monitored by current sensors, temperature sensors, vibration sensors, etc. to detect the status of motors, circuits and other key components. Environmental parameter sensors are used to monitor external conditions such as temperature, humidity, and light intensity to ensure that the goods in the vending machine are stored and sold under appropriate conditions. The data from these sensors are integrated and transmitted to the central control unit of the vending machine, which has efficient data processing capabilities, can generate analysis results based on the data collected by the sensors, and communicate bidirectionally with the remote management platform 102 through the communication module.

[0061] The communication module is an important part of the smart vending machine 101, which usually includes a wireless communication device, such as a Wi-Fi module, a 4G / 5G module or other suitable wireless communication equipment, and supports stable data exchange with the remote management platform 102. This module is used to receive various instructions sent by the remote management platform, such as commodity position adjustment instructions, fault warning signals or maintenance requests, and transmit the operation data of the vending machine to the remote management platform in real time for further analysis and management.

[0062] Smart vending machine 101 also integrates a customer service module to improve user experience. The module is equipped with a video call device, which enables customers to communicate with remote customer service personnel in real time. The video call device consists of a high-definition camera, microphone and speaker to ensure effective communication with customer service personnel. The smart customer service submodule is based on natural language processing technology and can automatically identify and answer customers' common questions, while guiding customers to manual customer service when necessary.

[0063] The design and structure of the smart vending machine 101 is intended to achieve efficient management of goods, flexible adjustment, fault prevention and intelligent user services, thereby comprehensively improving the operating efficiency and customer satisfaction of the vending machine.

[0064] Furthermore, the intelligent vending machine includes an environment adaptation module, and the environment adaptation module is specifically used for:

[0065] Real-time monitoring of environmental parameters around the vending machine through sensors, including temperature, humidity, light intensity and noise level;

[0066] Dynamically adjust the refrigeration system, lighting system and audio system inside the vending machine based on the monitored environmental parameters;

[0067] When the ambient noise level is detected to be high, the audio output volume and frequency range of the vending machine are adjusted to ensure that the user can clearly hear the voice prompts;

[0068] Adjust the brightness and contrast of the vending machine display according to changes in light intensity to optimize the user's visual experience;

[0069] Based on changes in ambient temperature and humidity, the storage temperature and humidity of different types of goods are intelligently adjusted to maintain optimal quality.

[0070] The smart vending machine includes an environmental adaptation module, which is designed to dynamically adjust the internal system according to changes in the vending machine's surrounding environment, thereby ensuring the quality of goods and the user experience. The core function of the environmental adaptation module is to achieve real-time monitoring and regulation of environmental parameters through a variety of sensors.

[0071] The environmental adaptation module is equipped with multiple high-precision sensors to continuously monitor the temperature, humidity, light intensity and noise level around the vending machine. The temperature sensor uses a semiconductor temperature sensing element, which can quickly respond to changes in the surrounding temperature to ensure the real-time and accuracy of the data. The humidity sensor is based on the principle of capacitive or resistive humidity detection and can effectively sense the humidity level in the air. The light intensity sensor is usually a photodiode or photoresistor, which can accurately measure the light intensity of the environment, while the noise sensor obtains real-time sound pressure change data in the environment through devices such as electret condenser microphones.

[0072] Environmental data collected by sensors are sent to the central processing unit of the vending machine for analysis. The environmental adaptation module dynamically adjusts the refrigeration, lighting and audio systems inside the vending machine based on changes in this data. For example, when the temperature sensor detects an increase in the outside temperature, the environmental adaptation module automatically increases the power of the refrigeration system in the vending machine to ensure that the temperature of the stored goods is not affected by the external high temperature, thereby maintaining the freshness and quality of the goods. If the outside humidity level decreases, the system will prevent the negative impact of the dry environment on certain products (such as fresh food or beverages) by adjusting the internal humidity control device.

[0073] When the noise sensor detects a high level of ambient noise, the environmental adaptation module automatically adjusts the output volume and frequency range of the vending machine's audio system. Specifically, the system will appropriately increase the volume of the voice prompts to overcome the interference of ambient noise and ensure that customers can clearly hear the voice prompts. In some cases, the module may also adjust the frequency range of the audio to focus the voice signal on a frequency band that is easier for the human ear to hear, in order to further improve the clarity of the voice prompts.

[0074] For the adjustment of the display system, the environmental adaptation module dynamically adjusts the brightness and contrast of the vending machine display by analyzing the data of the light intensity sensor. For example, when the light sensor detects that the external light is strong, the module will automatically increase the brightness and contrast of the display to avoid the screen content being difficult to identify due to strong light, ensuring the user's visual experience. In low-light environments, the system will reduce the brightness of the display to reduce energy consumption and prevent the strong light of the screen from irritating the user.

[0075] In addition, the environmental adaptation module also makes intelligent adjustments based on the storage requirements of different types of goods. For goods that need to be stored at low temperatures (such as cold drinks or fresh food), the module will accurately adjust the corresponding refrigeration temperature and humidity control equipment based on the feedback from the temperature and humidity sensors to ensure that the goods are stored under the best storage conditions. For goods stored at room temperature, the system will reasonably adjust the ventilation and temperature of the storage area based on the sensor data to avoid the impact of excessive humidity or temperature on the quality of the goods.

[0076] In short, the environmental adaptation module ensures the stable performance and user experience of the vending machine in different environments through real-time monitoring and dynamic adjustment, and can make detailed adjustments and optimizations to various environmental factors, thereby improving the overall intelligence level and operational effect of the vending machine.

[0077] Furthermore, the intelligent vending machine includes a dynamic pricing module, and the dynamic pricing module is used to:

[0078] Monitor surrounding environmental factors, including weather conditions, surrounding traffic, and nearby events;

[0079] Dynamically adjust commodity prices based on monitored environmental factors and commodity inventory conditions;

[0080] Launch promotional strategies under certain conditions;

[0081] Synchronize price adjustments and promotion information to vending machine displays and remote management platforms in real time.

[0082] Smart vending machines are equipped with a dynamic pricing module that dynamically adjusts the selling price of goods based on real-time environmental changes and product inventory. The module continuously monitors environmental factors around the vending machine, including weather conditions, surrounding traffic, and nearby events, through a variety of sensors and data input sources. For example, weather data can be obtained through networked meteorological information sources, traffic is monitored through the vending machine's built-in camera or infrared sensor, and information about nearby events can be obtained through the remote management platform.

[0083] After collecting this environmental data, the dynamic pricing module analyzes it and combines the analysis results with the current inventory level of the goods to determine the best pricing plan for the goods. For example, in hot weather, the module will increase the priority sales price of cold drinks; and when the flow of people increases significantly, the module may slightly increase the price of certain high-demand goods to balance supply and demand. When the inventory of a certain product is too high or close to expiration, the module will automatically trigger a price reduction strategy to accelerate the sales of the product through discounts or promotions. In this way, the system can ensure the circulation speed of goods while optimizing revenue.

[0084] The dynamic pricing module can also automatically initiate promotional strategies when environmental or inventory conditions meet specific trigger conditions. For example, when it detects that an upcoming event or holiday is about to begin, the module can offer limited-time discounts on popular products, or launch buy-one-get-one-free promotions for specific products during high-traffic periods. In addition, when the inventory level of the vending machine is below the safety line, the module will suspend promotional activities for some products to prevent inventory from running out.

[0085] The dynamic pricing module also has an information synchronization function. It will display price adjustments and promotion information on the vending machine screen in real time, and upload the adjustment results to the remote management platform. The vending machine display will dynamically update the price tags and promotional signs of the products according to the latest price changes and promotion strategies, so that customers can keep abreast of current discounts and offers. By receiving data from the dynamic pricing module, the remote management platform can monitor and analyze the overall pricing strategy, thereby further optimizing future pricing and promotion plans.

[0086] Through the implementation of the dynamic pricing module, smart vending machines can achieve flexible price adjustments and promotion management based on the surrounding environment and inventory status. This pricing mechanism enables vending machines to optimize sales strategies, improve commodity circulation efficiency, and maximize profits in various situations.

[0087] Furthermore, the intelligent vending machine includes a multimodal interaction module, and the multimodal interaction module is configured as follows:

[0088] Integrate speech recognition, gesture recognition and facial expression analysis technologies;

[0089] Allow users to interact with the vending machine through voice commands, gestures or facial expressions;

[0090] Adjust the vending machine’s response and tone based on the user’s interaction style and emotional state;

[0091] For users with limited mobility, the robot arm can be controlled by voice or gesture to take out the goods;

[0092] Record and analyze users' interaction preferences to personalize and optimize each user's shopping experience.

[0093] The smart vending machine is equipped with a multimodal interaction module to provide a more natural and user-friendly interaction method. The module integrates voice recognition, gesture recognition and facial expression analysis technologies, and is designed to allow users to interact with the vending machine in a variety of ways. Through the voice recognition function, users can give instructions to the vending machine in natural language, such as querying product information, adding products to the shopping cart, confirming purchases, and obtaining promotional information. The voice recognition module is equipped with a highly sensitive microphone and a dedicated voice processing chip, which can accurately capture and interpret the user's voice commands, and can maintain a high recognition accuracy even in a noisy environment.

[0094] Gesture recognition technology is implemented through cameras or infrared sensors installed on vending machines. It can detect the user's hand movements and convert them into corresponding operation instructions. Users can browse products, select products, add products to the shopping cart, and pay through specific gestures. For example, when a user points to a product, the vending machine automatically displays the detailed information of the product; if the user waves, the vending machine will think it is a signal to cancel the operation. The algorithm of the gesture recognition module can accurately recognize common gestures, and gradually improve its response speed and recognition accuracy through a deep learning model.

[0095] The facial expression analysis function is implemented through the camera installed on the vending machine, which can capture the user's facial features in real time and judge his or her current emotional state. The system analyzes the changes in the user's facial expression to determine the user's satisfaction, confusion or other emotional state, and adjusts the vending machine's feedback method accordingly. For example, when the vending machine detects that the user is confused, it will adjust the speed and content of the voice prompt to make it more detailed and easier to understand; when the user shows a happy emotion, the system may push promotional information in a more friendly tone.

[0096] For users with limited mobility, the multimodal interaction module provides special auxiliary functions. Users can control the vending machine's robotic arm through voice or gesture commands to complete the process of taking out the goods. This functional design helps ensure a barrier-free shopping experience. Users only need to give clear voice commands to the vending machine (such as "take out mineral water") or make corresponding gestures, and the robotic arm will automatically perform the pickup operation and deliver the goods to a location where the user can easily pick them up.

[0097] The multimodal interaction module also has the function of recording and analyzing interaction data. The system anonymizes and analyzes the user's voice, gesture and facial expression data to identify the user's interaction preferences and behavior patterns. Through the analysis of this data, the vending machine can gradually optimize the interaction experience with each user. For example, for users who frequently use voice commands, the system will give priority to providing a voice interaction interface; for users who are accustomed to gesture operation, the vending machine will speed up the gesture response speed. Personalized optimization not only improves the user experience, but also enhances the user stickiness and sales conversion rate of the vending machine.

[0098] Through the design of the above-mentioned multimodal interaction module, the smart vending machine can achieve more diversified and flexible user interaction methods, thereby significantly improving the user's shopping experience and convenience of use.

[0099] Furthermore, the smart vending machine includes a product expiration monitoring module, which periodically scans the expiration information of the products in the vending machine based on internal sensors and product label information; when it is detected that the product is about to expire, an alarm is sent to the management terminal through the remote management platform, and discount information or promotional activities are automatically generated according to the expiration date of the product to speed up the sales of the products that are about to expire.

[0100] Smart vending machines include a product expiration monitoring module, the main function of which is to track and manage the expiration information of the products in the vending machine in real time to ensure the quality of the products and optimize their sales. The product expiration monitoring module uses internal sensors and product label information to periodically scan the expiration date of the products. Each product is attached with a label when it is placed in the vending machine. These labels can be barcodes, QR codes or RFID tags, which store detailed information related to the product, including product name, production date, shelf life and expiration date.

[0101] The sensor in the product expiration monitoring module reads the expiration date information by scanning the product label. The scan can be performed periodically to ensure that the expiration date status of each product is updated in a timely manner. The module is also equipped with a data processing unit for analyzing and calculating the collected product expiration date information. The data processing unit compares the current time with the expiration date of the product and determines whether the product is nearing the end of its shelf life.

[0102] When the product expiration monitoring module detects that a product is about to expire, it will immediately generate an alarm message and send it to the management terminal through the remote management platform. This alarm message contains detailed information such as the name of the product, the remaining shelf life, and the inventory quantity, so that managers can take timely measures, such as manual inspection or replenishment.

[0103] In order to speed up the sales of goods that are about to expire, the product expiration monitoring module will automatically generate corresponding discount information or promotional activities. For example, when a batch of goods is only three days away from the expiration date, the system will automatically set the discount ratio of the goods, such as 10%, 20%, etc., and display the discount information to customers through the vending machine display. At the same time, the system can also adjust the priority display position of the goods in the vending machine according to the discount strategy, ensuring that customers can notice the promotional information of the goods at the first time, thereby increasing the sales speed of the goods.

[0104] In addition, the remote management platform will collect statistics and analyze the expired product information in the vending machine in order to adjust the future purchase and inventory strategy. This dynamic management based on the expiration date of the product not only improves the operating efficiency of the vending machine, but also reduces the losses caused by expired products.

[0105] The design of the product expiration monitoring module ensures the reliability of the smart vending machine in product quality management, and through automated discount generation and promotion strategies, it effectively speeds up the sales of products that are about to expire and avoids inventory backlogs.

[0106] The remote management platform 102 is configured to communicate bidirectionally with the smart vending machine via a wireless communication network. The remote management platform includes a product configuration module for generating product position adjustment instructions based on real-time sales data, promotional activities or customer behavior analysis. The product position adjustment instructions are sent to the smart vending machine to control the mechanical arm or movable shelf of the smart vending machine so that the vending machine can rearrange the products according to real-time needs; a fault prevention and self-repair module for analyzing sensor data in the vending machine using a machine learning algorithm to predict potential faults of mechanical components; when an abnormality is predicted, the vending machine is instructed to perform self-diagnosis and generate a repair request; a customer service module allows customers to communicate with remote customer service personnel in real time through the video call device on the vending machine to provide purchase recommendations, troubleshooting and return and exchange services; the customer service module also includes an intelligent customer service sub-module for responding to common problems of customers.

[0107] The remote management platform 102 is the core management system of the present invention, responsible for two-way communication and remote control with the intelligent vending machine 101. The platform establishes a real-time connection with the intelligent vending machine through a wireless communication network, can collect, analyze and process various data from the vending machine, and send instructions to realize intelligent management of commodity management, equipment maintenance and customer service.

[0108] The product configuration module of the remote management platform 102 generates product position adjustment instructions based on real-time collected sales data, promotional information and customer behavior analysis. Real-time sales data determines the distribution trend of hot-selling and slow-selling products by counting the sales frequency and customer preferences of the products in the vending machine; promotional information comes from the promotion management system in the background, which determines the priority display of products in a specific time period; customer behavior analysis uses sensor data in the vending machine, including the customer's line of sight, touch frequency, etc., to identify the customer's demand tendency. Based on these data, the product configuration module calculates the optimal product arrangement plan and sends the adjustment instructions to the smart vending machine 101 through the wireless communication network, driving the mechanical arm or movable shelf to rearrange the products to ensure that customers can obtain the required products more conveniently.

[0109] The remote management platform 102 also integrates a fault prevention and self-repair module to ensure the stable operation of the vending machine. The module continuously receives and monitors the sensor data of the vending machine, including information such as current, voltage, temperature, vibration and operating status. Through the built-in machine learning algorithm, the platform can identify potential fault signals, such as abnormal current, overheating, etc., and predict the failure risk of mechanical components based on historical data and trend analysis. When the fault prevention and self-repair module detects a possible fault, it automatically generates an early warning message and sends it to the smart vending machine 101 through the communication network, instructing it to perform self-diagnosis. At the same time, the module generates a repair request to notify the operation and maintenance personnel to conduct further inspection and maintenance.

[0110] The fault prevention and self-repair module in the remote management platform 102 is used to improve the operational reliability of the smart vending machine. To achieve this goal, the module continuously monitors the key operational data of the vending machine, including current, voltage, temperature, vibration, and the status of other mechanical components, through an integrated sensor network.

[0111] The machine learning algorithm is designed for the specific fault characteristics of smart vending machines. It is not just a simple fault classification or traditional anomaly detection, but a fault prevention based on multi-dimensional signal fusion and trend prediction. The following are the specific design and implementation details of the algorithm:

[0112] 1. Data preprocessing and feature extraction

[0113] The fault prevention and self-repair module first receives and integrates data from different sensors. Each sensor has a different signal type, such as:

[0114] Current and voltage sensors: Mainly used to monitor the status of the power system and determine whether the power supply is stable.

[0115] Temperature sensor: used to detect temperature changes inside the vending machine and its mechanical parts to determine whether there is overheating.

[0116] Vibration sensor: used to monitor the vibration patterns of robotic arms, shelves or other moving parts and identify abnormal vibration patterns.

[0117] Operation status sensor: includes signals such as displacement, speed, acceleration, etc. of mechanical parts to determine whether the mechanical parts are stuck or overloaded.

[0118] The data preprocessing process includes data standardization, noise filtering and feature extraction. The specific feature extraction steps are as follows:

[0119] Time series decomposition is performed on each sensor data to extract trend, seasonality and residual components.

[0120] According to the characteristics of different signals, the sliding window technology is used to generate the signal change rate in a short period of time (such as the slope change of current, the rate of increase of temperature, the frequency change of vibration, etc.).

[0121] Fourier transform is used to convert vibration signals into spectral features to detect potential fault frequency components of mechanical parts.

[0122] Wavelet transform is applied to perform multi-scale analysis on current and voltage signals to extract high-frequency and low-frequency components in order to identify short-term transient fluctuations and long-term trends.

[0123] 2. Failure risk prediction algorithm

[0124] In order to adapt to the fault characteristics of vending machines, this embodiment proposes a fault risk prediction algorithm based on a multi-dimensional time series fusion model. The specific design of the algorithm is as follows:

[0125] The algorithm adopts a multi-layer deep neural network (DNN) structure, which consists of multiple parallel long short-term memory (LSTM) units and convolutional neural network (CNN).

[0126] LSTM unit: used to process time series signals from different sensors. LSTM can memorize the historical state of sensor signals and identify abnormal change patterns, such as a continuous increase in temperature or a sudden increase in current.

[0127] CNN unit: used to process the spectral characteristics and local spatiotemporal characteristics of sensor signals, such as specific frequency components of the robot arm vibration signal and sudden changes in the voltage waveform.

[0128] Multi-dimensional fusion layer: The timing features and spectrum features of different signals are fused in a deep network to form a global fault risk indicator.

[0129] Model training:

[0130] The model is trained using self-supervised learning. Specifically, historical sensor data and actual fault records are used to generate a sample data set with fault labels. During the training process, the model not only learns the characteristics of different fault types, but also generates early warning patterns of potential faults through self-supervised signals.

[0131] Through multi-task learning, the model simultaneously learns the characteristics of different types of faults during training, including mechanical jams, power instability, and component overheating, ensuring the identification and prediction of multiple faults.

[0132] By adopting an incremental learning strategy, the model can be continuously updated and optimized when new fault data arrives, achieving rapid adaptation to different fault modes.

[0133] 3. Fault warning and self-repair

[0134] In actual operation, the algorithm calculates the fault risk score (RiskScore) in real time based on the sensor signal. Specifically, the algorithm outputs a risk score to assess the potential risk of the current fault, ranging from 0 to 1, with the closer to 1 indicating a higher fault risk.

[0135] When the risk score exceeds a certain set threshold (for example, 0.7), the fault prevention and self-repair module will automatically generate early warning information.

[0136] The warning information includes the possible fault type, location and estimated fault time.

[0137] This information is sent to the smart vending machine 101 via the communication network, instructing it to perform self-diagnosis. For example, the robot arm can verify whether there is a jamming problem by adjusting the motion trajectory; the refrigeration system can detect the response of the internal temperature sensor by short-term shutdown.

[0138] The module also generates a repair request to notify the operation and maintenance personnel to conduct inspection and maintenance. The repair request includes a detailed analysis of the cause of the failure, relevant sensor signal trend graphs, and possible solution suggestions.

[0139] The following is the reference implementation code of the machine learning algorithm in this embodiment:

[0140]

[0141]

[0142]

[0143]

[0144]

[0145] #Risk Score Prediction

[0146] with torch.no_grad():

[0147] risk_score=model(torch.tensor(processed_data,dtype=torch.float32).unsqueeze(0))

[0148] #If the risk score exceeds the threshold, generate warning information

[0149] if risk_score.item()>threshold:

[0150] warning_message = {

[0151] "Fault Type":"Possible overheating or mechanical jam",

[0152] "Occurrence location":"Vending machine 101",

[0153] "Estimated time": "Process immediately"

[0154] }

[0155] print("Warning:", warning_message)

[0156] repair_request={

[0157] "Cause Analysis": "The temperature sensor continues to be hot and the vibration signal is abnormal",

[0158] "Suggested measures":"Check the movement trajectory of the robot arm and repair the refrigeration system",

[0159] }

[0160] print("Repair request:",repair_request)

[0161] The customer service module is an important part of the remote management platform 102, which aims to enhance the customer's experience. The module is connected to the vending machine through a real-time video call device, allowing customers to communicate directly with remote customer service personnel. The video call function consists of a high-definition camera, microphone and speaker to ensure the clarity and stability of audio and video. Through video calls, customer service personnel can provide customers with purchase suggestions, answer troubleshooting questions or assist in returns and exchanges. In addition, the intelligent customer service submodule is based on natural language processing technology, which can automatically identify and process common customer questions, such as detailed information on products, promotional activities and how to use vending machines. When customers ask more complex questions, the intelligent customer service submodule can automatically transfer the conversation to manual customer service for seamless switching.

[0162] The remote management platform 102 also has powerful data analysis and management functions, and performs global monitoring and optimization of the operation of vending machines through the cloud data center. The platform aggregates the operation data of multiple vending machines and uses big data analysis technology to provide inventory trend forecasts, hot-selling product identification, and market demand analysis. These analysis results can support the optimal configuration of products and precision marketing strategies, thereby improving the sales performance and operational efficiency of vending machines.

[0163] Through the organic combination of the above modules and functions, the remote management platform 102 realizes remote control, real-time monitoring and intelligent maintenance of the smart vending machine, and provides efficient, flexible and intelligent remote management capabilities for the entire vending system.

[0164] Furthermore, the remote management platform includes a cross-selling optimization module, and the cross-selling optimization module is used to:

[0165] Analyze sales data and user purchasing behavior at different vending machine locations to identify correlations between products;

[0166] Generate personalized product combination recommendations for each vending machine based on the identified product associations;

[0167] Based on the generated personalized product combination recommendation, the movable shelf or robotic arm of the smart vending machine is guided to place the associated products in adjacent locations.

[0168] The cross-selling optimization module of the remote management platform is used to improve the combination sales effect of products. The module accurately identifies the correlation between products by analyzing sales data and user purchasing behavior at different vending machine locations. This correlation includes the products that customers often buy in combination in a single purchase, sales trends in a specific time period, and differences in preferences among customer groups. For example, the module may identify that in a vending machine at a certain location, customers tend to buy snacks at the same time as a beverage, or that during a specific promotion period, customers are more likely to buy a specific promotional combination at the same time.

[0169] In order to accurately identify the relevance of goods, the cross-selling optimization module relies on big data analysis and machine learning algorithms. The module obtains sales data from vending machines in real time, including the sales frequency, sales period, sales amount and customer purchase path of specific goods. At the same time, the analysis of user purchasing behavior is not limited to a single vending machine, but also covers the comparison of purchasing patterns between vending machines in different locations, thereby constructing a correlation matrix between goods. Through in-depth analysis of this matrix, the module can identify which goods have stronger purchase relevance at specific vending machine locations and determine the combination recommendation plan between these goods.

[0170] When generating personalized product combination recommendations, the cross-selling optimization module prioritizes products based on their relevance. This ranking not only considers the frequency of product combinations, but also takes into account customer preferences at specific vending machine locations. For example, in a vending machine in an office area, the module may prioritize a combination of coffee and energy bars, while in a vending machine near a school, it may prioritize a combination of juice and snacks. The personalized recommendation plan for each vending machine is constantly and dynamically adjusted based on real-time sales data and customer behavior to ensure the accuracy and effectiveness of product combination recommendations.

[0171] Once the personalized product combination recommendation is generated, the cross-selling optimization module will send the recommendation results to the central control unit of the vending machine to guide the movable shelf or robotic arm to make corresponding adjustments to the product positions. Specifically, the module will instruct the robotic arm or movable shelf to place highly related products in physically adjacent positions to facilitate customers to obtain multiple related products at one time. This position adjustment is based on an optimized path planning algorithm to ensure that the robotic arm can quickly and accurately rearrange the products during movement. At the same time, the layout adjustment will take into account the customer's behavioral habits and place commonly used combination products in a position that is easy for the customer to see and reach, thereby increasing the purchase rate of the combination products.

[0172] The design of this cross-selling optimization module not only improves the merchandise sales efficiency of the vending machine, but also improves the customer's shopping experience.

[0173] Furthermore, the remote management platform includes a group intelligence collaboration module, and the group intelligence collaboration module is used to:

[0174] Multiple smart vending machines with close geographical locations form a collaborative network;

[0175] Real-time analysis of inventory status and sales trends of each vending machine in the network;

[0176] When it is detected that the stock of a specific commodity in a designated smart vending machine is insufficient but the stock of a nearby smart vending machine is sufficient, a commodity allocation plan is generated;

[0177] According to the commodity allocation plan, unmanned delivery vehicles are called to allocate commodities.

[0178] The group intelligence collaboration module in the remote management platform is designed to improve the collaboration efficiency of multiple smart vending machines. This module connects multiple smart vending machines with similar geographical locations into a collaborative network to achieve unified management and resource sharing. Each smart vending machine maintains real-time connection with the group intelligence collaboration module through a wireless communication network, allowing the module to continuously monitor and analyze the inventory status and sales trends of each vending machine. Through this networked structure, data between different vending machines can be shared in real time, providing a basis for collaboration among vending machines in the region.

[0179] In daily operations, the swarm intelligence collaboration module automatically analyzes the inventory data of each vending machine to identify inventory differences and commodity circulation trends between vending machines. For example, if a certain beverage in a vending machine sells quickly and the inventory is rapidly decreasing, while the adjacent vending machine has sufficient inventory of the beverage, the module will immediately recognize this situation and mark it as a target that needs to be allocated. This analysis is based on a comprehensive consideration of the vending machine's geographical location, sales data, and commodity inventory information to ensure the rationality and effectiveness of the allocation decision.

[0180] When it is detected that a certain vending machine has insufficient inventory of a specific product, but a nearby vending machine has sufficient inventory, the swarm intelligence collaboration module will automatically generate a product allocation plan. The process of generating an allocation plan includes determining the allocation quantity, allocation path, and time schedule. The allocation plan will give priority to the vending machine with the shortest distance and sufficient inventory as the allocation source, while taking into account traffic conditions and distribution paths to ensure optimal allocation efficiency and cost.

[0181] Once the product allocation plan is determined, the group intelligent collaboration module will call the unmanned delivery vehicle through the network system to allocate the products. The unmanned delivery vehicle will extract the target products from the allocation source vending machine according to the set path and deliver them to the target vending machine with insufficient inventory. The unmanned delivery vehicle is equipped with a positioning system and an intelligent navigation module to ensure the safety and accuracy of the distribution process. The entire allocation process is monitored in real time by the group intelligent collaboration module, including the current location of the delivery vehicle, the distribution progress and the handover status of the goods, to ensure the smooth completion of the allocation task.

[0182] Through the deployment of group intelligent collaboration modules, the remote management platform can achieve coordinated deployment among multiple vending machines, which not only improves the supply balance of goods, but also reduces sales losses caused by insufficient inventory of individual vending machines. This collaborative mechanism significantly improves the operational efficiency and resource utilization of vending machines, ensuring that the supply of goods in the region is more flexible, fast and intelligent.

[0183] Furthermore, the remote management platform includes a social marketing and user participation module, and the social marketing and user participation module is configured as follows:

[0184] Generate a unique social media account for each smart vending machine;

[0185] Automatically publish information such as smart vending machine locations, inventory updates, promotions, etc. to social media platforms;

[0186] Allow users to interact with the vending machine through social media platforms, including pre-ordering items or participating in online voting to decide the next batch of stock;

[0187] Analyze user preferences and adjust product configuration based on social media interaction data;

[0188] Implement location-based social gamification marketing strategies, including punching in to unlock special offers or discounts.

[0189] The social marketing and user engagement module in the remote management platform aims to establish a closer and more interactive connection with users through social media. This module generates a unique social media account for each smart vending machine, enabling the vending machine to independently carry out marketing activities on social media platforms. In this way, each vending machine is not only an automatic vending terminal, but also a medium for social interaction.

[0190] The social marketing and user engagement module has an automatic publishing function, which can publish relevant information about the vending machine on social media platforms. This information includes the specific location of the vending machine, current inventory status, promotions and special offers. The update frequency of the published content is automatically adjusted according to the operating conditions of the vending machine. For example, when a certain product is low in inventory or is about to be sold out, the module will immediately generate a prompt message and publish it on social media to attract nearby users to buy it as soon as possible. At the same time, during specific promotional activities, the module can also generate promotional information related to the event, such as limited-time discounts, holiday specials, etc., to attract more users' attention and participation.

[0191] Users can interact with vending machines in various ways through social media platforms. For example, users can reserve products on social media to ensure that the products are still in stock when they arrive at the vending machine. In addition, the module also supports online voting, where users can vote on the next batch of products that need to be replenished in the vending machine. This participation mechanism based on user preferences not only enhances the user's sense of participation, but also effectively improves the rationality of the configuration and sales efficiency of the products in the vending machine.

[0192] The social marketing and user engagement module analyzes social media interaction data to identify user preference trends and behavior patterns. For example, by analyzing users' likes, comments, and votes on different products, the module can identify which products the current user group is more interested in, and dynamically adjust the product configuration in the vending machine accordingly. This real-time optimization mechanism based on user behavior enables vending machines to better adapt to market demand during operation, improve product sales efficiency and user satisfaction.

[0193] The module also implements social gamification marketing strategies through location positioning, such as encouraging users to clock in at specific vending machines to unlock special products or get discount rewards. The clock-in mechanism can be confirmed by the vending machine's positioning system. When the user completes the clock-in on social media, the system will generate corresponding discount information or product unlocking prompts. This social gamification interaction method not only enhances the user's entertainment experience, but also effectively attracts more users to the vending machine for consumption, thereby improving the overall operation effect of the vending machine.

[0194] Through social marketing and user engagement modules, smart vending machines not only have traditional sales functions, but also have the ability to establish continuous interaction with users.

[0195] Furthermore, the remote management platform includes an adaptive commodity layout and robotic arm control module, which realizes the optimal layout of commodities inside the vending machine through the following formula 1:

[0196]

[0197] Among them, L opt is the optimal layout solution set of the goods inside the vending machine; by maximizing the objective function in the formula, the system will determine the optimal product placement solution, thereby optimizing the location distribution of the goods. The objective function here refers to:

[0198]

[0199] m is the total number of commodities in the vending machine, indicating the number of all commodities that need to be optimized for layout.

[0200] W i is the weight coefficient of product i, reflecting the impact of the weight of the product on the operation of the robot arm; the weight of the product directly affects the difficulty and energy consumption of the robot arm. Heavier products may need to be placed in a position that is easy for the robot arm to carry, reducing the carrying path and operation time, thereby improving layout efficiency. i The value is usually based on the actual weight of the product, which is measured in advance by a sensor or weighing module.

[0201] B i is the average daily sales volume of product i; it indicates the average sales of a product over a period of time. This value can be obtained through historical sales data analysis. A higher average daily sales volume indicates that the product is more popular and should be placed in a location that is easy for customers to pick up.

[0202] B max The maximum average daily sales volume of the goods in the vending machine, that is, the sales volume of the goods with the highest average daily sales volume among all the goods in the vending machine. It is used to standardize the sales volume of each product so that the first term in the formula can relatively measure the sales performance of the product. This value is obtained by counting the historical sales data of all products.

[0203] G i The frequency of customer interaction with product i, including touch, gaze retention, and voice command. The interaction includes the number of times the customer touches the product, the time the customer stays at the product, and the number of voice commands. This value is monitored and counted in real time through multiple sensors in the vending machine (such as cameras, infrared sensors, and voice recognition modules).

[0204] G thIt is the threshold of interaction frequency, which is used to judge the interaction popularity of a product. This threshold can be set according to the product type and historical interaction data.

[0205] H i is the current shelf height position of product i, indicating the actual placement height of the product in the vending machine.

[0206] H max It is the maximum shelf height in the vending machine, representing the highest placement position available in the vending machine.

[0207] α,μ,v,ρ i To adjust the coefficient, it can be set directly through experimental data technology or according to expert knowledge.

[0208] After calculating the optimal layout solution set L opt After that, according to L opt Generate a rearrangement instruction for the goods in the vending machine; and control the mechanical arm to adjust the position of the goods according to the rearrangement instruction.

[0209] The process of obtaining the optimal layout solution set mainly involves the following key steps:

[0210] First, the system needs to collect and organize comprehensive data on all the products in the vending machine. This includes factors such as product weight, sales volume, customer interaction frequency, current shelf height, etc. The system will standardize this information to ensure that different factors are comparable in the subsequent optimization process.

[0211] Next, the system calculates a "layout score" for each product based on its performance in the vending machine. This score is a comprehensive evaluation based on multiple factors, including the product's sales frequency, customer interaction, weight, current location, etc. The layout score of each product reflects its relative importance and adaptability in the vending machine. The system evaluates the layout scores of all products in different locations through iterative calculations.

[0212] After calculating the placement score for each product, the system compares the scores of all products and tries to find a globally optimal placement. To achieve this, the system uses an optimization algorithm that tries many different combinations of products and evaluates the overall score of each combination. The optimization algorithm constantly adjusts the position of the products with the goal of maximizing the overall score.

[0213] Finally, after multiple iterations and optimizations, the system will come up with a product arrangement with the highest overall score, which is the "optimal layout solution set". This solution set defines the specific location of each product in the vending machine. The system will then convert the solution into specific instructions and send them to the robotic arm in the vending machine, which will rearrange the products according to the instructions.

[0214] Furthermore, the remote management platform includes an integrated inventory and pricing optimization module, which simultaneously calculates the optimal inventory configuration set S of the goods through the following formula 2: opt and the optimal price set P opt :

[0215]

[0216] Among them, S opt is the optimal inventory configuration set, which represents the optimal inventory configuration set of each commodity in the vending machine, S opt ={S1,S2,…S i …,S n}, where S i represents the optimal inventory configuration of the i-th product in the vending machine; P opt is the optimal commodity price set, P opt = {P1,P2,…P i …,P n}, where S i represents the optimal price of the i-th product in the vending machine; n is the total number of products in the vending machine;

[0217] C s The product shortage cost coefficient is used to measure the cost of lost sales opportunities due to insufficient inventory. This coefficient can be obtained through historical sales data analysis, such as potential losses when inventory is insufficient.

[0218] D i is the historical demand for commodity i, which indicates the average sales volume of the commodity in a certain period of time in the past. It can be obtained through the sales records of vending machines and used to evaluate the market demand for commodities.

[0219] I i is the current inventory of product i, indicating the current inventory level of the product in the vending machine. It is monitored and updated in real time through sensors in the vending machine or inventory management system.

[0220] C h The commodity holding cost coefficient is used to reflect the holding costs caused by excessive inventory of commodities, such as the cost of commodity storage and space occupation. This value can be calculated based on the storage cost and historical holding data of the commodity.

[0221] I i is the current inventory of product i, indicating the current inventory level of the product in the vending machine. It is monitored and updated in real time through sensors in the vending machine or inventory management system.

[0222] I max is the maximum inventory capacity of commodity i, indicating the maximum quantity of the commodity that can be stored in the vending machine. This parameter is usually set according to the design specifications of the vending machine or the characteristics of the commodity.

[0223] E i is the number of days until product i expires; E th The threshold number of days to process expiring products in advance;

[0224] D i is the historical demand for commodity i, obtained through historical data;

[0225] D max The maximum demand for commodity i refers to the highest sales volume of commodity i in history within a period of time (such as a day, a week, or a month). It reflects the highest demand peak of the commodity in the past sales record and is used to evaluate the maximum possible sales of the commodity during the busiest period.

[0226] w1, w2, λ, λ1, κ, and κ1 are adjustment coefficients, which are obtained by calculation from experimental data or directly set based on expert knowledge;

[0227] For the i-th product in the vending machine, according to the calculated optimal inventory configuration S i , generate a replenishment plan and carry out replenishment;

[0228] According to the calculated optimal price P i , adjust the price of goods in the vending machine and update it in real time on the vending machine display.

[0229] Get the optimal inventory configuration set S opt and the optimal price set P opt The implementation steps mainly include the following steps.

[0230] First, the system needs to collect key data for each product in the vending machine, including historical demand, current inventory, maximum inventory capacity, number of days until the product expires, and other product-related parameters. This data comes from sensors and sales records inside the vending machine to ensure that it can fully and accurately reflect the status and demand of the product.

[0231] After data collection and collation, the system enters the optimization calculation phase. This phase involves calculating inventory configuration and price adjustments for each product at the same time. To achieve this goal, the system evaluates the impact of multiple factors for each product, such as shortage cost, holding cost, demand level, inventory status, expiration risk, etc. These factors are analyzed comprehensively so that the system can balance between different goals.

[0232] Next, the system will perform multiple rounds of iterative optimization. In each round of iteration, the system will try different inventory and pricing combinations and calculate the effect of each combination to determine whether it helps optimize inventory configuration and pricing. In this process, the system will continuously adjust the inventory volume and price of the goods and evaluate the impact of these adjustments on overall inventory and sales. As the iterations proceed, the system gradually finds an optimal combination that can maximize the balance between inventory holding, shortages, and pricing benefits.

[0233] After multiple optimization iterations, the system will eventually determine an overall optimal solution, namely the optimal inventory configuration set S opt and the optimal price set P opt Among them, S opt represents the optimal inventory level of each commodity, and P opt is the optimal price for each product.

[0234] In getting S opt After that, the system will automatically generate a replenishment plan, including the replenishment quantity, time and priority of each product to ensure that the inventory is maintained at the optimal level. opt Automatically adjust the selling price of items and update pricing information in real time on the vending machine display to ensure pricing is consistent with current inventory and market demand.

[0235] Through the above process, the system can dynamically adjust the inventory and price in the vending machine to maximize inventory efficiency and sales revenue.

[0236] Although the present application is disclosed as above in the form of a preferred embodiment, it is not intended to limit the present application. Any technical personnel in this field may make possible changes and modifications without departing from the spirit and scope of the present application. Therefore, the scope of protection of the present application shall be based on the scope defined by the claims of the present application.

Claims

1. An intelligent vending machine and remote management system, characterized in that: include: Intelligent vending machines, which include movable shelves or mechanical arms for storing and rearranging merchandise positions; Sensors for monitoring the inventory status of goods inside the vending machine, the operating status of mechanical components, and environmental parameters; and a communication module for communicating with a remote management platform; A remote management platform is configured to communicate bidirectionally with the smart vending machine via a wireless communication network. The remote management platform includes a product configuration module for generating product position adjustment instructions based on real-time sales data, promotional activities or customer behavior analysis. The product position adjustment instructions are sent to the smart vending machine to control the mechanical arm or movable shelf of the smart vending machine so that the vending machine can rearrange the products according to real-time needs; a fault prevention and self-repair module is used to analyze the sensor data in the vending machine using a machine learning algorithm to predict potential failures of mechanical components; when an abnormality is predicted, the vending machine is instructed to perform self-diagnosis and generate a repair request; a customer service module allows customers to communicate with remote customer service personnel in real time through the video call device on the vending machine to provide purchase recommendations, troubleshooting and return and exchange services; the customer service module also includes an intelligent customer service sub-module for responding to common customer questions.

2. The intelligent vending machine and remote management system according to claim 1, characterized in that: The intelligent vending machine includes an environment adaptation module, which is specifically used for: Real-time monitoring of environmental parameters around the vending machine through sensors, including temperature, humidity, light intensity and noise level; Dynamically adjust the refrigeration system, lighting system and audio system inside the vending machine based on the monitored environmental parameters; When the ambient noise level is detected to be high, the audio output volume and frequency range of the vending machine are adjusted to ensure that the user can clearly hear the voice prompts; Adjust the brightness and contrast of the vending machine display according to changes in light intensity to optimize the user's visual experience; Based on changes in ambient temperature and humidity, the storage temperature and humidity of different types of goods are intelligently adjusted to maintain optimal quality.

3. The intelligent vending machine and remote management system according to claim 1, characterized in that: The remote management platform includes a cross-selling optimization module, which is used to: Analyze sales data and user purchasing behavior at different vending machine locations to identify correlations between products; Generate personalized product combination recommendations for each vending machine based on the identified product associations; Based on the generated personalized product combination recommendation, the movable shelf or robotic arm of the smart vending machine is guided to place the associated products in adjacent locations.

4. The intelligent vending machine and remote management system according to claim 1, characterized in that: The smart vending machine includes a dynamic pricing module, which is used to: Monitor surrounding environmental factors, including weather conditions, surrounding traffic, and nearby events; Dynamically adjust commodity prices based on monitored environmental factors and commodity inventory conditions; Launch promotional strategies under certain conditions; Synchronize price adjustments and promotion information to vending machine displays and remote management platforms in real time.

5. The intelligent vending machine and remote management system according to claim 1, characterized in that: The remote management platform includes a group intelligence collaboration module, and the group intelligence collaboration module is used to: Multiple smart vending machines with close geographical locations form a collaborative network; Real-time analysis of inventory status and sales trends of each vending machine in the network; When it is detected that the stock of a specific commodity in a designated smart vending machine is insufficient but the stock of a nearby smart vending machine is sufficient, a commodity allocation plan is generated; According to the commodity allocation plan, unmanned delivery vehicles are called to allocate commodities.

6. The intelligent vending machine and remote management system according to claim 1, characterized in that: The intelligent vending machine includes a multimodal interaction module, and the multimodal interaction module is configured as follows: Integrate speech recognition, gesture recognition and facial expression analysis technologies; Allow users to interact with the vending machine through voice commands, gestures or facial expressions; Adjust the vending machine’s response and tone based on the user’s interaction style and emotional state; For users with limited mobility, the robot arm can be controlled by voice or gesture to take out the goods; Record and analyze users' interaction preferences to personalize and optimize each user's shopping experience.

7. The intelligent vending machine and remote management system according to claim 1, characterized in that: The remote management platform includes a social marketing and user participation module, and the social marketing and user participation module is configured as follows: Generate a unique social media account for each smart vending machine; Automatically publish information such as smart vending machine locations, inventory updates, promotions, etc. to social media platforms; Allow users to interact with the vending machine through social media platforms, including pre-ordering items or participating in online voting to decide the next batch of stock; Analyze user preferences and adjust product configuration based on social media interaction data; Implement location-based social gamification marketing strategies, including punching in to unlock special offers or discounts.

8. The intelligent vending machine and remote management system according to claim 1, characterized in that: The smart vending machine includes a product expiration monitoring module, which regularly scans the expiration information of the products in the vending machine based on internal sensors and product label information; when it is detected that the product is about to expire, an alarm is sent to the management terminal through the remote management platform, and discount information or promotional activities are automatically generated according to the expiration date of the product to speed up the sales of the products that are about to expire.

9. The intelligent vending machine and remote management system according to claim 1, characterized in that: The remote management platform includes an adaptive commodity layout and robotic arm control module, which realizes the optimal layout of commodities inside the vending machine through the following formula 1: Among them, L opt is the optimal layout solution set of goods inside the vending machine; m is the total number of goods in the vending machine; W i is the weight coefficient of commodity i, reflecting the impact of the commodity weight on the operation of the robot arm; B i is the average daily sales volume of commodity i; B max is the maximum average daily sales volume of the goods in the vending machine; G i is the customer's interaction frequency with product i, including touch, gaze, and voice command; G th is the threshold of interaction frequency; H i is the current shelf height position of product i; H max is the maximum shelf height in the vending machine; α,μ,ν,ρ i is the adjustment coefficient; After calculating the optimal layout solution set L opt After that, according to L opt Generate a rearrangement instruction for the goods in the vending machine; and control the mechanical arm to adjust the position of the goods according to the rearrangement instruction.

10. The intelligent vending machine and remote management system according to claim 1, characterized in that: The remote management platform includes an integrated inventory and pricing optimization module, which simultaneously calculates the optimal inventory configuration set S of the goods through the following formula 2: opt and the optimal price set P opt : Among them, S opt is the optimal inventory configuration set, which represents the optimal inventory configuration set of each commodity in the vending machine, S opt ={S1,S2,…S i …,S n }, where S i represents the optimal inventory configuration of the i-th product in the vending machine; P opt is the optimal commodity price set, P opt = {P1,P2,…P i …,P n }, where S i represents the optimal price of the i-th product in the vending machine; n is the total number of products in the vending machine; C s is the commodity shortage cost coefficient; D i is the historical demand for commodity i; I i is the current inventory of commodity i; C h I is the commodity holding cost coefficient; i is the current inventory of product i; max is the maximum inventory capacity of commodity i; E i is the number of days until product i expires; E th The threshold number of days to process expired products in advance; D i is the historical demand for commodity i; D max is the maximum demand for commodity i; w1, w2, λ, λ1, κ, κ1 are adjustment coefficients; For the i-th product in the vending machine, according to the calculated optimal inventory configuration S i , generate a replenishment plan and carry out replenishment; According to the calculated optimal price P i , adjust the price of goods in the vending machine and update it in real time on the vending machine display.

Images