Block chain enabled electronic product recovery traceability management platform and method

Abstract

The invention discloses a block chain enabled electronic product recovery traceability management platform and method, and relates to the technical field of traceability management platforms, after each electronic product recovery behavior is completed, user behavior characteristics and recovery point characteristics are recorded and written into a block chain, different user behavior characteristics and different recovery point characteristics are combined, and the traceability management platform is established. The method comprises the following steps: generating a plurality of state perceptions, establishing a perception set for the plurality of state perceptions, sorting all state perceptions in the perception set based on a traceability factor, adapting an excitation strategy for each state perception through an upper confidence bound algorithm, writing an adaptation result into a block chain for storage, and when the perception set changes according to a scene, sending the adaptation result to the block chain for storage. And dynamically updating the adaptation result in the block chain. The management system constructs a state perception set and introduces an upper confidence bound algorithm to automatically match an optimal incentive strategy for each state perception, can adjust an incentive mode in real time according to a recovery environment and user characteristics, realizes personalized accurate incentive, and greatly improves the user participation degree and the recovery efficiency.

Description

Technical Field

[0001] The present invention relates to the technical field of traceability management platforms, and in particular to a blockchain-enabled electronic product recycling traceability management platform and method. Background Art

[0002] In the current context of green sustainable development and circular economy, the issue of electronic product recycling and reuse has received increasing attention. Electronic products are updated and replaced at a rapid pace, and the number of discarded electronic devices (such as mobile phones, computers, home appliances, etc.) is increasing. If not handled properly, it will not only cause waste of resources, but also pose a serious threat to the environment and human health. In order to improve the recycling rate of electronic products, standardize the recycling process and achieve traceable management, it is urgent to establish an efficient, transparent and reliable recycling management mechanism.

[0003] The existing technology has the following deficiencies: 1. Traditional recycling incentive strategies are mostly based on fixed rules and cannot be intelligently adjusted based on different user behaviors and real-time conditions at recycling points. This results in generally poor incentive effects. For example, some users respond enthusiastically to point-based incentives, while others may prefer physical redemption or ranking-driven incentives. However, existing systems often adopt a unified incentive model and fail to achieve precise incentives tailored to individual needs. This results in wasted incentive resources and insufficient user motivation to recycle. 2. Existing incentive mechanisms lack a systematic learning and optimization process, and the incentive effects cannot be quantified and continuously improved. In most systems, once the incentive strategy is set, it becomes fixed for a long time and lacks the ability to dynamically optimize based on changes in user behavior. Moreover, the effectiveness of the incentive often depends on subjective judgment or later feedback, lacking quantifiable real-time evaluation methods and intelligent update mechanisms. This static strategy design is difficult to adapt to the complex needs of changes in user groups, regional differences, and adjustments to operational goals.

[0004] Based on this, this application proposes a blockchain-enabled electronic product recycling traceability management platform and method. By constructing a state perception set and introducing an upper confidence bound algorithm to automatically match the optimal incentive strategy for each state perception, the incentive method can be adjusted in real time according to the recycling environment and user characteristics, achieving personalized and precise incentives, and greatly improving user participation and recycling efficiency. Summary of the Invention

[0005] The purpose of the present invention is to provide a blockchain-enabled electronic product recycling traceability management platform and method to address the deficiencies in the background technology.

[0006] To achieve the above objectives, the present invention provides the following technical solutions: a blockchain-enabled electronic product recycling traceability management platform and method, the management method comprising the following steps: After each electronic product recycling activity is completed, the user behavior characteristics and recycling point characteristics are recorded and written into the blockchain; Combine different user behavior characteristics and different recycling point characteristics to generate several state perceptions, establish a perception set from these state perceptions, and sort all state perceptions in the perception set based on the traceability factor; The upper confidence bound algorithm is used to adapt the incentive strategy for each state perception, and the adaptation results are written into the blockchain storage. When the perception set changes according to the scenario, the adaptation results in the blockchain are dynamically updated.

[0007] In a preferred embodiment, the incentive strategy is adapted for each state perception through the upper confidence bound algorithm, and the adaptation results are written to the blockchain storage, including the following steps: Multiple incentive strategy options are preset to form a strategy set, and historical adaptation benefit information is recorded for each strategy, including the number of times each incentive strategy was selected and the average benefit of each strategy in historical selections; Based on the current state perception, the upper confidence bound algorithm is used to calculate the upper confidence bound value of each incentive strategy under the state perception; After calculating the confidence upper bound values ​​for all incentive strategies, the incentive strategy corresponding to the maximum confidence upper bound value is selected for current state perception and adapted.

[0008] In a preferred embodiment, multiple incentive strategy options are preset, including point rewards, cash back and carbon point subsidies, which constitute a strategy set ,in represents the jth incentive strategy, and k represents the number of incentive strategies.

[0009] In a preferred embodiment, based on the current state perception, an upper confidence bound algorithm is used to calculate the confidence upper bound value of each incentive strategy under the state perception. The calculation expression is: ,in: Incentive strategy The upper confidence bound of Incentive strategy The historical average return of Indicates the current total state perception number, Incentive strategy The number of times it has been used, It represents the adjustment coefficient for adjusting the balance between exploration and exploitation.

[0010] In a preferred embodiment, when the perception set changes according to the scenario, dynamically updating the adaptation results in the blockchain includes the following steps: Monitor scenario parameters related to electronic product recycling behavior. When any one or more parameters exceed the preset threshold, it automatically identifies changes in the current recycling scenario and triggers a dynamic update process. Update the state perception information of the current perception set in real time, collect user characteristic information of current recycling behavior and environmental characteristics of recycling points; After data collection is completed, the recycling frequency and equipment failure rate of each state perception are recalculated, and the upper confidence bound algorithm is reapplied to each state perception to adapt the optimal incentive strategy; The calculated incentive strategy adaptation results for each state perception are written into the blockchain. The update operation adopts the structured block data writing method, while retaining the historical adaptation version to form the on-chain incentive strategy evolution trajectory.

[0011] In a preferred embodiment, scenario parameters related to electronic product recycling behavior are monitored, including changes in the flow of people at the recycling point, changes in user delivery frequency, fluctuations in recycling trends for different device categories, queue time, and incentive policy response rate.

[0012] In a preferred embodiment, sorting all state perceptions in the perception set based on the traceability factor includes the following steps: Obtain the recovery frequency and failure rate of each state perception in the perception set, normalize the recovery frequency and failure rate so that their value ranges are mapped to [0,1], sum the normalized recovery frequency and failure rate to obtain the traceability factor, and sort all state perceptions from large to small according to the traceability factor.

[0013] In a preferred embodiment, different user behavior characteristics and different recycling point characteristics are combined to generate a plurality of state perceptions, and the plurality of state perceptions are combined to form a perception set, including the following steps: Read each stored recycling record in the blockchain, which contains user behavior characteristics and recycling point characteristics; On the basis of each record, the two types of features are associated and combined to construct state perception; Aggregate multiple state perceptions in the same time period or geographical area to form a perception set; When constructing the perception set, tag information is added to each state perception, including the region to which it belongs, the recycling type, the time tag, and the working status of the equipment.

[0014] In a preferred embodiment, after each electronic product recycling behavior is completed, the user behavior characteristics and recycling point characteristics are recorded, including the following steps: After each electronic product recycling activity is completed, the information collection process is automatically triggered to record the recycling-related data and classify the data; Collect user behavior characteristics, including the user's unique identity, the number of the recycling device used, the delivery time, the weight of the electronic product delivered, and the type of delivered item; Extract the environmental and operational characteristics of the current recycling point, namely the recycling point characteristics. The recycling point characteristics include the physical geographical location of the recycling equipment, the real-time or periodic traffic level at the equipment location, and the current average waiting time in queues.

[0015] The blockchain-enabled electronic product recycling and traceability management platform includes a feature acquisition module, a perception set establishment module, a strategy adaptation module, and a dynamic update module; Feature acquisition module: After each electronic product recycling behavior is completed, the user behavior characteristics and recycling point characteristics are recorded and written into the blockchain; Perception set building module: combines different user behavior characteristics and different recycling point characteristics to generate several state perceptions, and then builds a perception set from these state perceptions; Strategy Adaptation Module: After sorting all state perceptions in the perception set based on the traceability factor, the incentive strategy is adapted for each state perception using the upper confidence bound algorithm; Dynamic update module: When the perception set changes according to the scenario, the adaptation results in the blockchain are dynamically updated.

[0016] In the above technical solution, the technical effects and advantages provided by the present invention are: After each electronic product recycling activity is completed, the present invention records user behavior characteristics and recycling point characteristics and writes them to the blockchain. This system then combines different user behavior characteristics with different recycling point characteristics to generate several state perceptions, which are then organized into a perception set. Within the perception set, all state perceptions are sorted based on traceability factors. An upper confidence bound algorithm is used to adapt an incentive strategy to each state perception, and the adaptation results are stored on the blockchain. When the perception set changes based on the scenario, the adaptation results in the blockchain are dynamically updated. This management system constructs a state perception set and incorporates an upper confidence bound algorithm to automatically match the optimal incentive strategy for each state perception. This system can adjust incentives in real time based on the recycling environment and user characteristics, achieving personalized and precise incentives, significantly improving user engagement and recycling efficiency. Dynamic updates based on scenario changes enable closed-loop learning, continuous optimization, and result verification of recycling incentive schemes, thus establishing a self-evolving intelligent recycling management system. BRIEF DESCRIPTION OF THE DRAWINGS

[0017] In order to more clearly illustrate the embodiments of the present application or the technical solutions in the prior art, the following briefly introduces the drawings required for use in the embodiments. Obviously, the drawings described below are only some embodiments described in the present invention. For ordinary technicians in this field, other drawings can also be obtained based on these drawings.

[0018] Figure 1 The figure is a flow chart of the management method of the present invention.

[0019] Figure 2 This is a system architecture diagram of the management system of the present invention. DETAILED DESCRIPTION

[0020] To make the objectives, technical solutions, and advantages of the embodiments of the present invention more clear, the technical solutions in the embodiments of the present invention will be clearly and completely described below in conjunction with the accompanying drawings of 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. All other embodiments obtained by ordinary technicians in this field based on the embodiments of the present invention without making any creative efforts shall fall within the scope of protection of the present invention.

[0021] Example 1: Please refer to Figure 1 As shown, the blockchain-enabled electronic product recycling traceability management method described in this embodiment includes the following steps: After each electronic product recycling behavior is completed, the user behavior characteristics and recycling point characteristics are recorded and written into the blockchain to ensure that the information cannot be tampered with and is traceable. User behavior characteristics include user ID, device ID, delivery time, weight, and type, and recycling point characteristics include location, traffic flow, queue time, etc. Different user behavior characteristics and different recycling point characteristics are combined to generate several state perceptions, and several state perceptions are established into a perception set. After all state perceptions are sorted in the perception set based on the traceability factor, the incentive strategy is adapted for each state perception through the upper confidence bound algorithm (UCB), and the adaptation results are written into the blockchain storage. When the perception set changes according to the scenario, the adaptation results in the blockchain are dynamically updated.

[0022] This application records user behavior characteristics and recycling point characteristics after each electronic product recycling activity is completed and writes them to the blockchain. It then combines different user behavior characteristics with different recycling point characteristics to generate several state perceptions, which are then combined into a perception set. Within the perception set, all state perceptions are sorted based on traceability factors. An upper confidence bound algorithm is used to adapt an incentive strategy for each state perception, and the adaptation results are written to the blockchain for storage. When the perception set changes based on the scenario, the adaptation results in the blockchain are dynamically updated. This management system constructs a state perception set and introduces an upper confidence bound algorithm to automatically match the optimal incentive strategy for each state perception. It can adjust the incentive method in real time based on the recycling environment and user characteristics, achieving personalized and precise incentives, significantly improving user engagement and recycling efficiency. Dynamic updates based on scenario changes enable closed-loop learning, continuous optimization, and result verification of the recycling incentive scheme, thus building an intelligent recycling management system with self-evolving capabilities.

[0023] The dynamic effects of this application are as follows: Recycling points are switched from urban cores to remote areas: Changes manifested: reduced traffic in remote areas and high equipment idleness; Strategic impact: The MAB algorithm may increase the frequency of selecting "high incentive arms" (such as cashback or lottery); System feedback: Increase recycling volume in the region through stronger incentives; Change: The incentive strategy arm chooses options that are biased towards "high reward".

[0024] When the user type changes, for example, from an environmentalist to an ordinary resident: Change manifestations: The proportion of new users or low-activity users increased; Strategy impact: The system increases the probability of exploration and tries more incentive combinations; System feedback: Identify incentives that are more suitable for new users; Changes: The strategy gradually adapts to the new user structure and stabilizes recycling behavior.

[0025] In cross-regional operation and promotion scenarios: Changes: Multiple new regions launched, with diverse user profiles; Strategic impact: The system increases exploratory capabilities and quickly locates the optimal strategy for each region; System feedback: forming a “strategy profile” in different regions; Change: Changed to a multi-environment parallel adaptation mode (can evolve into Contextual-Bandit).

[0026] Example 2: After each electronic product recycling activity is completed, the user behavior characteristics and recycling point characteristics are recorded and written into the blockchain to ensure that the information cannot be tampered with and is traceable. The user behavior characteristics include user ID, device ID, delivery time, weight, and type, etc. The recycling point characteristics include location, traffic volume, queue time, etc., including the following steps: After each electronic product recycling operation is completed, the system automatically triggers an information collection process, comprehensively recording key data related to that recycling operation and categorizing the data. The first thing collected is user behavior characteristics, which primarily include the user's unique identifier (such as user ID), the number of the recycling device used (device ID), the specific time of delivery, the weight of the electronic product delivered, and the type of item delivered (such as mobile phones, computers, and household appliances). This information collectively describes the user's behavior trajectory and the characteristics of the items in this recycling operation, providing data support for subsequent analysis of user recycling habits, delivery frequency, and carbon emission reduction contributions.

[0027] At the same time, the system also extracts information about the current recycling point's environmental and operational characteristics, known as recycling point characteristics. These include, but are not limited to, the physical location of the recycling equipment (down to GPS coordinates or street information), the real-time or periodic foot traffic levels at the equipment's location (e.g., during peak hours, midday, and on holidays), and the current average queue wait time (reflecting the equipment's load). These characteristics collectively constitute the spatial and temporal context of recycling behavior and are crucial for determining whether recycling behavior is representative and whether strategies require local adaptation.

[0028] Once collected, all of this information is structured into standardized data records, representing the complete data entity for a single recycling transaction. To ensure data security, authenticity, and non-repudiation during storage and transmission, the system uses a cryptographic hashing algorithm to digest the data and digitally signs it to ensure it has not been tampered with before or after it is written. This data entity is then written to the blockchain, becoming an immutable historical record on the blockchain's distributed ledger.

[0029] Through blockchain's distributed consensus mechanism and chain-structured storage, each recycling record can be traced back to its entire creation process and the data sources it relies on. This not only enhances the transparency and credibility of recycling data, but also provides fundamental data support for subsequent scenarios such as carbon asset trading, policy subsidy issuance, green point rewards, and environmental behavior supervision. Furthermore, the entire process eliminates the need for third-party data verification. The system automatically establishes trust through algorithms and blockchain collaboration, thereby building a decentralized, highly reliable, and sustainable data management framework for electronics recycling.

[0030] Combining different user behavior characteristics and different recycling point characteristics to generate several state perceptions, and then establishing a perception set from these state perceptions; The process of combining different user behavior characteristics and recycling point characteristics to generate several state perceptions aims to extract and construct state information that can be used for decision support from multiple dimensions and scenarios to support the matching and optimization of intelligent incentive strategies. This process includes the following steps: First, the system reads each stored recycling record from the blockchain. This record contains specific user behavior characteristics (such as user ID, delivery time, device ID, delivery type, and delivery weight) and recycling point characteristics (such as geographic location, traffic density, queue time, and device status). Based on each record, the system correlates and combines these two types of characteristics to construct a state unit, which is a "state perception." State perception is essentially a comprehensive representation of a recycling behavior in a specific time and space scenario, reflecting the dynamic interaction between users and devices during the recycling process.

[0031] Next, the system uses strategies such as sliding time windows, regional clustering, or feature-weighted clustering to aggregate multiple state perceptions within the same time period or geographic area to form a "perception set." This perception set summarizes state information about recycling scenarios within a certain range. It covers the behavioral interactions of different users at multiple recycling points and serves as the basis for multi-arm incentive selection and strategy optimization.

[0032] When constructing the perception set, the system also adds tagging information to each state perception, such as the region, recycling type, time stamp, and device operating status, for subsequent classification and screening. To improve subsequent processing efficiency, the state perceptions in the perception set can be grouped or indexed according to preset dimensions, such as classification by recycling type or sorting by device load. This allows the system to quickly locate the most representative state subset when executing policy matching.

[0033] Through the above process, the system can dynamically and comprehensively reflect the current operational status and user activity within the electronics recycling network. The construction of a perception set not only improves the pertinence and adaptability of policy matching but also provides a solid data foundation for subsequent traceability factor evaluation, incentive strategy selection, and behavior optimization. This mechanism greatly enhances the recycling system's ability to perceive complex environments, enabling it to adjust strategies in a timely manner as user behavior and recycling equipment status change, thereby improving overall operational efficiency and user participation.

[0034] Assume that in a city’s electronics recycling network, there are three smart recycling points located at: Point A: office area; Point B: Entrance to the residential area; Point C: Inside the university campus; During the afternoon hours (e.g., 2:00 PM to 4:00 PM), the system collected the following three pieces of recycling behavior data, one from each user: Example of recycling behavior data: User behavior characteristics (behavior 1); User ID: U1001; Equipment ID: A01 (recycling point A); Delivery time: 14:23; Delivery weight: 1.2 kg; Delivery Type: Used laptop computers; Recycling point characteristics (Point A) Geographical location: office area (CBD); Current traffic volume: high (approximately 20 people per minute); Waiting time: 4 minutes; User behavior characteristics (behavior 2) User ID: U2033; Equipment ID: B12 (recycling point B); Delivery time: 15:01; Delivery weight: 0.8 kg; Delivery type: Used mobile phones; Recycling point characteristics (point B) Geographical location: Residential area; Current traffic: medium (approximately 8 people per minute); Waiting time: 1 minute; User behavior characteristics (behavior 3) User ID: U3450; Equipment ID: C07 (recycling point C); Delivery time: 15:40; Delivery weight: 2.5 kg; Delivery type: used monitors; Recycling point characteristics (point C); Geographical location: University area; Current traffic: low (approximately 3 people per minute); Waiting time: None; State perception generation process: The system combines each set of user behavior characteristics with the corresponding recycling point characteristics to generate three state perception records, as follows: State perception S1: {user U1001 + device A01 + high traffic + high waiting time + old computer + 1.2kg}; State perception S2: {user U2033 + device B12 + medium flow + short waiting time + used mobile phone + 0.8kg}; State perception S3: {user U3450 + device C07 + low traffic + no waiting + waste monitor + 2.5kg}; Schematic diagram of perception set construction: The system builds a perception set based on the current time window (e.g. 14:00–16:00), which contains state perceptions S1, S2, and S3. The perception set can be organized as follows: Classification by geographical distribution: CBD, university, and residential; Grouped by recycling type: laptops, mobile phones, monitors; Sort by device load: high queue time → low queue time; Application significance: The perception set will be used for subsequent sorting based on traceability factors, allocation of UCB incentive strategies, and dynamic adaptation of user behavior reward models.

[0035] For example, although S3 has a zero waiting time, it has a large delivery volume and is a high-value device; the system may give priority to point incentives to increase regional participation in universities.

[0036] Although the delivery weight of S1 is low, it is located in a CBD area with dense traffic and busy equipment. Therefore, the system may choose to issue green behavior badges to guide user behavior shifts during peak hours.

[0037] Sorting all state perceptions in the perception set based on the traceability factor includes the following steps: Obtain the recovery frequency and failure rate of each state perception in the perception set, normalize the recovery frequency and failure rate so that their value ranges are mapped to [0,1], sum the normalized recovery frequency and failure rate to obtain the traceability factor, and sort all state perceptions from large to small according to the traceability factor.

[0038] Counts the total number of recycling records for the current recycling point within a set time window (such as the past 24 hours or 7 days). Recycling frequency refers to how frequently a recycling point corresponding to a certain state perception is used by users within a specific time frame. It is a key indicator for measuring the representativeness and data density of recycling behavior. Recycling frequency is calculated by dividing the total number of recycling records by the time window (recording duration).

[0039] The failure rate reflects the frequency of failures in a certain recycling device per unit time and is a core indicator for evaluating data quality and equipment stability. The failure rate is calculated by counting the number of failures that occur at the current recycling point within a set time window and dividing the number of failures by the time window (recording duration).

[0040] The larger the traceability factor value, the more frequent the state perception and the more failures occur, which means it should be given priority attention. Sorting all state perceptions in the perception set from high to low according to the traceability factor value will help: control blockchain storage resources and filter out low-value data; accurately optimize behavioral incentives and recycling policies; and quickly locate potential pollution sources or responsible nodes.

[0041] The incentive strategy is adapted for each state perception through the upper confidence bound algorithm (UCB), and the adaptation results are written to the blockchain storage, including the following steps: The system presets multiple incentive strategy options, such as point rewards, cash back, carbon point subsidies, etc., to form a strategy set ,in represents the jth incentive strategy, k represents the number of incentive strategies, and at the same time, records the historical adaptation benefit information for each strategy, including: the number of times each strategy is selected , the average return of each strategy in historical choices , these data can be estimated and updated through the responses generated by users’ past recycling behaviors (such as increased activity, increased recycling volume, etc.).

[0042] Based on the current state perception, the upper confidence bound (UCB) algorithm is used to calculate the selection score of each incentive strategy under the state perception. The formula is as follows: ,in: Incentive strategy The upper confidence bound of Incentive strategy The historical average return of Indicates the current total number of state perceptions (equivalent to the number of system iterations), Incentive strategy The number of times it has been used, Represents the adjustment coefficient (usually 1 or 2) for adjusting the balance between exploration and exploitation, calculated for all strategies After the value, select The corresponding strategy serves as an incentive scheme for current state perception.

[0043] For each state perception, a policy adaptation record is constructed, including: State-aware unique identifier (such as user ID, timestamp, recycling point ID); Matching incentive policy identifier (e.g., "points + 5%"); Current traceability factor value; The corresponding upper confidence bound value; The above data is written into the blockchain in the form of a structured contract to ensure: the traceability of the strategy allocation process, the immutability of strategy parameters and allocation basis, and the reliability of the basis for subsequent auditing, evaluation and tuning.

[0044] When the perception set changes according to the scenario, the adaptation results in the blockchain are dynamically updated, including the following steps: When the perception set changes based on the scenario, the system needs to dynamically update the incentive strategy adaptation results stored in the blockchain in a timely manner to ensure that the incentive mechanism continues to be targeted and effective. This dynamic update process not only enhances the system's adaptive capabilities, but also ensures the continuous linkage between data traceability and behavioral incentives in complex environments. The entire update process includes the following key steps, each of which revolves around "change detection - information reconstruction - strategy re-adaptation - on-chain update". The details are as follows: First, the system continuously monitors key scenario parameters related to electronic product recycling behavior. These parameters include, but are not limited to, changes in foot traffic at recycling points, changes in user drop-off frequency, fluctuations in recycling trends for different device categories, continued increases or decreases in queue times, and significant decreases in the response rate to incentive strategies. When any one or more parameters exceed the preset fluctuation threshold, the system automatically identifies that the current recycling scenario has changed. For example, if a recycling point experiences a surge in users within a short period of time, if the device categories suddenly become homogenous, or if a historically effective strategy suddenly becomes ineffective, the system will identify this as a "scenario switch event," triggering the subsequent dynamic update process.

[0045] Secondly, the system updates the state perception information in the current perception set in real time. At this point, the system recollects user characteristics of the current recycling behavior (such as user ID, delivery time, device type, and weight), as well as the latest environmental characteristics of the recycling point (such as location, real-time traffic density, and changes in queue time). After data collection is complete, the system recalculates the recycling frequency and device failure rate for each state perception and normalizes these updated indicators to ensure that all state perceptions are compared on the same numerical scale. This reflects the traceability value and priority of the state in the current scenario.

[0046] The system then reorders all state perceptions based on the newly generated traceability factor, with the larger the traceability factor, the higher the ranking. This ranking reflects the system's priority for different states in the new environment and forms the basis for subsequent reallocation of incentive strategies. This ranking not only determines the order in which incentive strategies are allocated resources but also provides a rational basis for subsequent strategy selection, avoiding wasting resources on inefficient or low-impact states.

[0047] After ranking, the system reapplies the upper confidence bound (UCB) algorithm for each state perception to select the optimal incentive strategy. The UCB algorithm now combines historical average returns (i.e., the historical effectiveness of an incentive strategy in that state) with the current number of uses, and combines this with the time step to perform an exploration-utilization balance calculation. The system tends to increase the probability of selecting strategies that have been used less frequently but have greater potential in the current environment, thereby promoting strategy innovation and evolution. At the same time, the system appropriately lowers the scores of strategies that have performed well in the past but are currently experiencing a decline in response rate, prompting the system to more quickly escape local optima.

[0048] Ultimately, the system writes the recalculated incentive strategy adaptation results for each state perception to the blockchain. Updates typically utilize structured block data writing, while preserving historical adaptation versions to form an on-chain track of the incentive strategy's evolution. Updates include the newly calculated traceability factor, new ranking position, new incentive strategy identifier, the current perception scenario label, and the corresponding timestamp and version number. Because blockchains are tamper-proof and fully traceable, this update process ensures transparent and traceable policy changes, effectively supporting subsequent policy evaluation and accountability audits.

[0049] Furthermore, after each adaptation update, the system also creates an evolutionary track record of the incentive strategy allocation across multiple versions. This record can be used for subsequent big data training, traceability analysis, and automatic prediction of future scenarios. For example, if a strategy performs well in multiple similar scenarios, the system can pre-label it as a "recommended strategy." Conversely, if a strategy's performance consistently deteriorates across multiple environments, it can be labeled a "pending strategy."

[0050] In summary, this dynamic update mechanism realizes the stability, adaptability and intelligent evolution capability of the electronic product recycling and traceability system in a complex dynamic environment through change perception, data update, strategy re-adaptation and on-chain writing throughout the entire process, providing solid technical support for achieving precise incentives and traceability management throughout the entire life cycle.

[0051] Example 3: Please refer to Figure 2 As shown, the blockchain-enabled electronic product recycling traceability management platform described in this embodiment includes a feature acquisition module, a perception set establishment module, a policy adaptation module, and a dynamic update module; Feature acquisition module: After each electronic product recycling activity is completed, the user behavior characteristics and recycling point characteristics are recorded and written into the blockchain to ensure that the information cannot be tampered with and is traceable. User behavior characteristics include user ID, device ID, delivery time, weight, and type. Recycling point characteristics include location, traffic flow, queue time, etc. User behavior characteristics and recycling point characteristics are sent to the perception set establishment module; Perception set establishment module: combines different user behavior characteristics and different recycling point characteristics to generate several state perceptions, establishes a perception set from these state perceptions, and sends the perception set to the policy adaptation module; Strategy Adaptation Module: After sorting all state perceptions in the perception set based on the traceability factor, the upper confidence bound (UCB) algorithm is used to adapt the incentive strategy for each state perception, and the adaptation results are sent to the dynamic update module; Dynamic update module: When the perception set changes according to the scenario, the adaptation results in the blockchain are dynamically updated.

[0052] It should be understood that the term "and / or" as used herein simply describes a relationship between associated objects, indicating that three possible relationships exist. For example, "A and / or B" can represent: A alone, A and B together, or B alone. A and B can be singular or plural. Furthermore, the character " / " as used herein generally indicates an "or" relationship between the associated objects, but it may also indicate an "and / or" relationship. For specific understanding, please refer to the context.

[0053] It should be understood that in the various embodiments of the present application, the size of the serial numbers of the above-mentioned processes does not mean the order of execution. The execution order of each process should be determined by its function and internal logic, and should not constitute any limitation on the implementation process of the embodiments of the present application.

[0054] Those skilled in the art will appreciate that the units and algorithm steps of each example described in conjunction with the embodiments disclosed herein can be implemented in electronic hardware, or a combination of computer software and electronic hardware. Whether these functions are performed in hardware or software depends on the specific application and design constraints of the technical solution. Professional and technical personnel can use different methods to implement the described functions for each specific application, but such implementation should not be considered to be beyond the scope of this application. Those skilled in the art will clearly understand that, for the convenience and brevity of description, the specific working processes of the systems, devices and units described above can refer to the corresponding processes in the aforementioned method embodiments and will not be repeated here.

[0055] The above description is merely a specific embodiment of the present application, but the scope of protection of the present application is not limited thereto. Any changes or substitutions that can be easily conceived by a person skilled in the art within the technical scope disclosed in this application should be included in the scope of protection of this application. Therefore, the scope of protection of this application should be based on the scope of protection of the claims.

Claims

1. The blockchain-enabled electronic product recycling traceability management method is characterized by: The management approach includes the following steps: After each electronic product recycling activity is completed, the user behavior characteristics and recycling point characteristics are recorded and written into the blockchain; Combine different user behavior characteristics and different recycling point characteristics to generate several state perceptions, establish a perception set from these state perceptions, and sort all state perceptions in the perception set based on the traceability factor; The upper confidence bound algorithm is used to adapt the incentive strategy for each state perception, and the adaptation results are written into the blockchain storage. When the perception set changes according to the scenario, the adaptation results in the blockchain are dynamically updated.

2. The blockchain-enabled electronic product recycling traceability management method according to claim 1, characterized in that: The incentive strategy is adapted for each state perception through the upper confidence bound algorithm, and the adaptation results are written to the blockchain storage, including the following steps: Multiple incentive strategy options are preset to form a strategy set, and historical adaptation benefit information is recorded for each strategy, including the number of times each incentive strategy was selected and the average benefit of each strategy in historical selections; Based on the current state perception, the upper confidence bound algorithm is used to calculate the upper confidence bound value of each incentive strategy under the state perception; After calculating the confidence upper bound values ​​for all incentive strategies, the incentive strategy corresponding to the maximum confidence upper bound value is selected for current state perception and adapted.

3. The blockchain-enabled electronic product recycling traceability management method according to claim 2, characterized in that: Preset multiple incentive strategy options, including point rewards, cash back and carbon point subsidies, forming a strategy set ,in represents the jth incentive strategy, and k represents the number of incentive strategies.

4. The blockchain-enabled electronic product recycling traceability management method according to claim 3 is characterized by: Based on the current state perception, the upper confidence bound algorithm is used to calculate the confidence upper bound value of each incentive strategy under the state perception. The calculation expression is: ,in: Incentive strategy The upper confidence bound of Incentive strategy The historical average return of Indicates the current total state perception number, Incentive strategy The number of times it has been used, It represents the adjustment coefficient for adjusting the balance between exploration and exploitation.

5. The blockchain-enabled electronic product recycling traceability management method according to claim 4 is characterized by: When the perception set changes according to the scenario, the adaptation results in the blockchain are dynamically updated, including the following steps: Monitor scenario parameters related to electronic product recycling behavior. When any one or more parameters exceed the preset threshold, it automatically identifies changes in the current recycling scenario and triggers a dynamic update process. Update the state perception information of the current perception set in real time, collect user characteristic information of current recycling behavior and environmental characteristics of recycling points; After data collection is completed, the recycling frequency and equipment failure rate of each state perception are recalculated, and the upper confidence bound algorithm is reapplied to each state perception to adapt the optimal incentive strategy; The calculated incentive strategy adaptation results for each state perception are written into the blockchain. The update operation adopts the structured block data writing method, while retaining the historical adaptation version to form the on-chain incentive strategy evolution trajectory.

6. The blockchain-enabled electronic product recycling traceability management method according to claim 5 is characterized by: Monitor scenario parameters related to electronic product recycling behavior, including changes in the flow of people at the recycling point, changes in user delivery frequency, fluctuations in recycling trends for different device categories, queue times, and response rates to incentive strategies.

7. The blockchain-enabled electronic product recycling traceability management method according to claim 6, characterized in that: Sorting all state perceptions in the perception set based on the traceability factor includes the following steps: Obtain the recovery frequency and failure rate of each state perception in the perception set, normalize the recovery frequency and failure rate so that their value ranges are mapped to [0,1], sum the normalized recovery frequency and failure rate to obtain the traceability factor, and sort all state perceptions from large to small according to the traceability factor.

8. The blockchain-enabled electronic product recycling traceability management method according to claim 7, characterized in that: Combining different user behavior characteristics and different recycling point characteristics to generate several state perceptions, and establishing a perception set from these state perceptions, includes the following steps: Read each stored recycling record in the blockchain, which contains user behavior characteristics and recycling point characteristics; On the basis of each record, the two types of features are associated and combined to construct state perception; Aggregate multiple state perceptions in the same time period or geographical area to form a perception set; When constructing the perception set, tag information is added to each state perception, including the region to which it belongs, the recycling type, the time tag, and the working status of the equipment.

9. The blockchain-enabled electronic product recycling traceability management method according to claim 8, characterized in that: After each electronic product recycling activity is completed, the user behavior characteristics and recycling point characteristics are recorded, including the following steps: After each electronic product recycling activity is completed, the information collection process is automatically triggered to record the recycling-related data and classify the data; Collect user behavior characteristics, including the user's unique identity, the number of the recycling device used, the delivery time, the weight of the electronic product delivered, and the type of delivered item; Extract the environmental and operational characteristics of the current recycling point, namely the recycling point characteristics. The recycling point characteristics include the physical geographical location of the recycling equipment, the real-time or periodic traffic level at the equipment location, and the current average waiting time in queues.

10. A blockchain-enabled electronic product recycling traceability management platform, used to implement the management method described in any one of claims 1 to 9, characterized in that: It includes feature acquisition module, perception set establishment module, strategy adaptation module and dynamic update module; Feature acquisition module: After each electronic product recycling behavior is completed, the user behavior characteristics and recycling point characteristics are recorded and written into the blockchain; Perception set building module: combines different user behavior characteristics and different recycling point characteristics to generate several state perceptions, and then builds a perception set from these state perceptions; Strategy Adaptation Module: After sorting all state perceptions in the perception set based on the traceability factor, the incentive strategy is adapted for each state perception using the upper confidence bound algorithm; Dynamic update module: When the perception set changes according to the scenario, the adaptation results in the blockchain are dynamically updated.

Images