Forestry industry scheduling optimization method based on multi-source data

Abstract

This invention discloses a forestry industry scheduling optimization method based on multi-source data, belonging to the field of industry scheduling technology. This invention integrates remote sensing, logging measurement, transfer identification, weighbridge, inventory age, production shifts, demand planning, road and meteorological data, and energy consumption and carbon budget data to construct an event-aligned dataset and a compliance constraint set. It incorporates logging quotas, logging ban windows, quarantine routes, load limits, and energy consumption and carbon budgets into a solvable scheduling optimization model, and introduces inventory baselines and quality decay penalties to achieve unified optimization of reducing inventory, minimizing inventory age loss, and prioritizing supply. It outputs auditable scheduling schemes and vouchers, and can be rolled over and rearranged based on feedback deviations.

Description

Technical Field

[0001] This invention relates to the field of industrial scheduling technology, and in particular to a forestry industry scheduling optimization method based on multi-source data. Background Technology

[0002] The forestry industry chain typically spans administrative resource units, distribution centers, processing plants, and key engineering timber supply ends, and is constrained by administrative rules such as logging quotas, logging bans, quarantine routes, road load limits, and energy consumption and carbon budgets. Current management relies heavily on annual quota allocation and manual coordination. Data such as remote sensing of forest stands, logging operation measurement, timber transfer identification, weighbridge measurements, inventory age, processing capacity shifts, demand planning, road capacity, and weather risks are scattered across different systems, making it difficult to form a unified event-aligned dataset for use in a single solvable model. This often results in problems such as inventory backlog and increasing inventory age leading to quality degradation, supply demand and industry operations mutually constraining each other, inability to quickly reorganize under sudden road closures and weather risks, and difficulty in auditing and verifying decision-making basis, affecting the timeliness, compliance, and traceability of administrative scheduling.

[0003] Currently, Chinese invention patent application number 202211536009.5 discloses a method for optimizing load scheduling in agricultural industrial parks considering carbon trading. The method includes: Step 1: Obtaining system carbon emission quotas and constructing a carbon emission cost model; Step 2: Analyzing the load adjustability characteristics of agricultural industrial parks and establishing a corresponding mathematical model; Step 3: Establishing an objective function that minimizes system operating costs and carbon trading costs; Step 4: Constructing constraints such as thermoelectric power equality constraints and agricultural industrial park load time-shifting constraints; Step 5: Solving the objective function to obtain an optimized load scheduling scheme for agricultural industrial parks. Based on considering the carbon emissions of agricultural systems and the load adjustability characteristics of agricultural industrial parks, this method analyzes and establishes a corresponding mathematical model for agricultural multi-energy complementary systems. Through simulation examples, an optimized scheduling scheme considering the time-shifting nature of agricultural industrial park loads under a carbon trading mechanism is obtained. This method can coordinate the economic and low-carbon aspects of a thermoelectric multi-energy complementary agricultural system.

[0004] The relevant technologies struggle to simultaneously meet compliance constraints such as logging quotas and ban windows, quarantine routes, road capacity, and energy consumption and carbon budgets, to construct a solvable scheduling optimization model with the core objectives of reducing inventory and decreasing storage age decay, and to output an auditable cross-entity scheduling solution. Summary of the Invention

[0005] The technical problem solved by this invention is that existing technologies are unable to construct a solvable scheduling optimization model with the core objectives of reducing inventory and reducing inventory decay, and output an auditable cross-entity scheduling scheme, under the premise of simultaneously satisfying compliance constraints such as logging quotas and ban windows, quarantine routes, road capacity, and energy consumption carbon budgets.

[0006] To solve the above-mentioned technical problems, the present invention provides the following technical solution: The forestry industry scheduling optimization method based on multi-source data includes the following steps: Step S1: Collect remote sensing forest stand data, forest compartment ownership data, logging operation measurement data, timber transfer identification data, weighbridge measurement data, inventory age data, processing capacity shift data, demand planning data, road traffic capacity data, meteorological risk data, and energy consumption carbon budget data; clean and spatiotemporally align the data to generate an event-aligned dataset. Step S2: Generate a set of compliance constraints based on the event alignment dataset; Step S3: Calculate inventory status data and quality decay penalty data, and establish an inventory baseline; Step S4: Generate available supply forecast data and administrative demand forecast data; Step S5: Input the compliance constraint set, inventory baseline, and forecast data into the scheduling optimization model to solve, and output the scheduling scheme data. Step S6: Generate execution instruction data and audit voucher data based on the scheduling scheme data and update them on a rolling basis.

[0007] Preferably, step S1 includes the following sub-steps: Step S101: Obtain remote sensing forest stand data, forest compartment ownership data, logging operation measurement data, timber transfer identification data, weighbridge measurement data, inventory age data, processing capacity shift data, demand planning data, road traffic capacity data, meteorological risk data, and energy consumption carbon budget data. Step S102: Perform missing data removal, outlier correction and unit consistency processing on the data collected in step S101, and output cleaned multi-source data. Step S103: Perform time synchronization on the cleaned multi-source data based on a unified time reference, and output time synchronization data; Step S104: Based on the sub-compartment boundaries in the forest compartment ownership data, spatial and node assignments are performed on the time synchronization data, and spatially aligned data is output. Step S105: Aggregate the spatially aligned data and organize it by period to output the event aligned dataset.

[0008] Preferably, step S104 includes: Administrative resource units are determined based on forest compartment ownership data, and remote sensing forest stand data and logging operation measurement data are assigned to the corresponding administrative resource units. Based on inventory age data, processing capacity shift data, and weighbridge measurement data, industry nodes are determined, and timber transfer identification data is assigned to the corresponding industry nodes. Based on road capacity data, the connectivity relationships and capacity parameters of road segments are determined, and the reachability relationships between industry nodes are written into the event alignment dataset.

[0009] Preferably, step S2 includes the following sub-steps: Step S201: Analyze the ownership data of forest compartments and extract logging quota parameters, prohibited logging area markers, prohibited logging window parameters, and quarantine risk area markers; Step S202: Analyze the road capacity data and extract the road segment load limit parameters, control window parameters, and maximum capacity parameters; Step S203: Analyze the energy consumption carbon budget data and extract the processing energy consumption budget parameters and transportation carbon budget parameters; Step S204: Organize the logging quota parameters, prohibited logging window parameters, quarantine risk zone identification, road section load limit parameters, control window parameters, processing energy consumption budget parameters, and transportation carbon budget parameters into compliance constraint parameters, and output the compliance constraint set.

[0010] Preferably, step S204 includes: Convert the logging quota parameter into a quota constraint that the logging volume does not exceed the quota; Convert the prohibited mining zone identifier and prohibited mining window parameters into a window constraint where the logging volume for the prohibited mining period is zero; Converting quarantine risk zone markers into timber transfer marker data requires path constraints through designated industry nodes; The road segment load limit parameters and maximum traffic capacity parameters are converted into traffic constraints that ensure the transport volume does not exceed the road segment capacity. The processing energy consumption budget parameters and transportation carbon budget parameters are converted into budget constraints that ensure that the energy consumption and carbon emissions of each period do not exceed the budget. A rule basis number is generated for each constraint and written into the compliance constraint set.

[0011] Preferably, step S3 includes the following sub-steps: Step S301: Based on inventory age data, timber transfer identification data and weighbridge measurement data, calculate the inbound quantity, outbound quantity and ending inventory quantity according to industry nodes and periods, and output inventory status data. Step S302: Calculate the inventory age index based on the inventory age data, and associate the inventory age index with the inventory status data; Step S303: Based on meteorological risk data, the inventory age index is exposed and corrected to generate quality decay penalty data, which characterizes the intensity of loss caused by the increase of inventory age. Step S304: Generate an inventory baseline based on historical inventory status data and quality decay penalty data.

[0012] Preferably, step S4 includes the following sub-steps: Step S401: Calculate the recoverable resources based on remote sensing forest stand data and forest compartment ownership data, and calibrate them in conjunction with logging operation measurement data to output basic supply data. Step S402: Based on meteorological risk data and road capacity data, accessibility and risk reduction are performed on the basic supply data, and the available supply forecast data is output. Step S403: Generate administrative demand forecast data based on demand planning data, and calculate the demand urgency index. The demand urgency index is jointly determined by the delivery window, supply guarantee level, and default impact item. Step S404: Write the demand urgency index into the administrative demand forecast data to characterize the supply guarantee priority.

[0013] Preferably, step S5 includes the following sub-steps: Step S501: Establish a scheduling optimization model and set the logging scheduling quantity, transportation scheduling quantity, allocation scheduling quantity and processing and feeding scheduling quantity as decision objects; Step S502: The optimization objectives are to minimize the ending inventory, minimize the loss corresponding to the quality degradation penalty data, minimize the delay penalty of the administrative demand forecast data, and minimize the transportation and processing costs. Step S503: The compliance constraint set is used as a hard constraint, the inventory baseline, inventory status data, processing capacity shift data and road traffic capacity data are used as conservation and capacity constraints, and the available supply forecast data and administrative demand forecast data are used as supply and demand boundary constraints. Step S504: Solve the scheduling optimization model and output the scheduling scheme data, which includes phased logging plans, transportation plans, allocation plans, processing plans and delivery plans.

[0014] Preferably, step S504 includes: When the scheduling optimization model is a linear programming model, the linear programming solution strategy is used to output the scheduling scheme data. When the scheduling optimization model contains discrete constraints such as logging shifts or vehicle shifts, the scheduling optimization model is constructed as a mixed integer linear programming model and a branch and bound solution strategy is used to output scheduling scheme data. If the compliance constraint set makes the model infeasible, the hard constraints corresponding to the rule basis number remain unchanged, and the administrative demand forecast data with low demand urgency index is subject to controllable degradation, and the scheduling scheme data containing degradation records is output.

[0015] Preferably, step S6 includes the following sub-steps: Step S601: Generate logging execution instruction data, transportation execution instruction data, allocation execution instruction data, and processing scheduling execution instruction data based on the scheduling scheme data, and send them to the management terminal; Step S602: Collect execution feedback data based on timber transfer identification data, weighbridge measurement data and logging operation measurement data, write the execution feedback data into the event alignment dataset and update the inventory status data and inventory age data. Step S603: Generate audit voucher data, which includes a data reference list, a rule basis number list, constraint satisfaction records, and inventory reduction comparison records. Step S604: When the deviation of the returned data from the scheduling scheme data exceeds the preset deviation threshold or the meteorological risk data triggers the preset risk condition, a rolling update is triggered and steps S2 to S6 are re-executed.

[0016] The beneficial effects of this invention are as follows: This invention integrates remote sensing, logging measurement, transfer identification, weighbridge, inventory age, production shifts, demand planning, road and meteorological data, and energy consumption and carbon budget data to construct an event-aligned dataset and a set of compliance constraints. It incorporates logging quotas, logging ban windows, quarantine routes, load limits, and energy consumption and carbon budgets into a solvable scheduling optimization model, and introduces inventory baselines and quality decay penalties to achieve unified optimization of reducing inventory, minimizing inventory age losses, and prioritizing supply. It outputs auditable scheduling schemes and vouchers, and can be rolled over based on feedback deviations. Attached Figure Description

[0017] Figure 1 A flowchart illustrating the steps of a forestry industry scheduling optimization method based on multi-source data, provided in one embodiment of the present invention. Detailed Implementation

[0018] To make the above-mentioned objects, features and advantages of the present invention more apparent and understandable, the specific embodiments of the present invention will be described in detail below with reference to the accompanying drawings. Obviously, the described embodiments are only some embodiments of the present invention, and not all embodiments.

[0019] Example, refer to Figure 1 This paper provides a forestry industry scheduling optimization method based on multi-source data, including the following steps: Step S1: Collect remote sensing forest stand data, forest compartment ownership data, logging operation measurement data, timber transfer identification data, weighbridge measurement data, inventory age data, processing capacity shift data, demand planning data, road traffic capacity data, meteorological risk data, and energy consumption carbon budget data, clean and spatiotemporally align them to generate an event-aligned dataset.

[0020] Step S2: Generate a set of compliance constraints based on the event alignment dataset.

[0021] Step S3: Calculate inventory status data and quality degradation penalty data, and establish an inventory baseline.

[0022] Step S4: Generate available supply forecast data and administrative demand forecast data.

[0023] Step S5: Input the compliance constraint set, inventory baseline, and forecast data into the scheduling optimization model to solve, and output the scheduling scheme data.

[0024] Step S6: Generate execution instruction data and audit voucher data based on the scheduling scheme data and update them on a rolling basis.

[0025] This invention integrates remote sensing, logging measurement, transfer identification, weighbridge data, inventory age, production shifts, demand planning, road and meteorological data, and energy and carbon budget data to construct an event-aligned dataset and a set of compliance constraints. It incorporates logging quotas, logging ban windows, quarantine routes, load limits, and energy and carbon budgets into a solvable scheduling optimization model, and introduces inventory baselines and quality decay penalties to achieve unified optimization of reducing inventory, minimizing inventory age losses, and prioritizing supply. It outputs auditable scheduling schemes and vouchers, and can be rolled over based on feedback deviations.

[0026] This embodiment addresses a scenario where provincial forestry authorities coordinate the dispatch of several state-owned forest farms, distribution centers, and processing plants. The dispatch cycle is divided into several periods per week. The goal is to minimize the ending inventory and reduce quality degradation losses due to inventory aging, while meeting compliance constraints such as logging quotas, prohibited logging windows, quarantine routes, road capacity, and energy consumption and carbon budgets. Simultaneously, it aims to reduce penalties for delays in administrative demand forecasting data and generate auditable supporting documentation.

[0027] Step S1 includes the following sub-steps: Step S101: Obtain remote sensing forest stand data, forest compartment ownership data, logging operation measurement data, timber transfer identification data, weighbridge measurement data, inventory age data, processing capacity shift data, demand planning data, road traffic capacity data, meteorological risk data, and energy consumption carbon budget data.

[0028] Remote sensing forest stand data should at least include canopy coverage, canopy height, vegetation index, change detection results, and sampling time within the sub-compartment boundaries; sub-compartment ownership data should at least include sub-compartment boundaries, ownership entity, forest type, prohibited logging area markers, prohibited logging window parameters, and logging quota parameters; logging operation measurement data should at least include operation team identifier, operation location, operation start time, operation end time, logging volume measurement value, and corresponding sub-compartment identifier; timber transfer identification data should at least include single log or batch identifier, source sub-compartment identifier, warehousing node identifier, warehousing node identifier, transfer timestamp, and quarantine status identifier; weighbridge measurement data should at least include weighing node identifier, weighing timestamp, vehicle number identifier, gross weight, tare weight, net weight, and corresponding timber transfer identifier; inventory age data should at least include industry node identifier, timber... Material flow identification, warehousing timestamp, current inventory, storage age in days, and stacking method identification; processing capacity shift data should at least include processing node identification, period identification, number of available shifts, unit shift capacity, and unit output energy consumption; demand planning data should at least include demand entity identification, delivery node identification, period identification, demand quantity, delivery window, and supply guarantee level; road traffic capacity data should at least include road segment identification, starting node identification, ending node identification, period identification, maximum traffic capacity parameter, road segment load limit parameter, and control window parameter; meteorological risk data should at least include period identification, node or area identification, rainfall, number of consecutive rainy days, extreme temperature index, and risk level; energy consumption carbon budget data should at least include processing node identification, period identification, processing energy consumption budget parameter, and transportation carbon budget parameter.

[0029] Step S102: Perform missing data removal, outlier correction, and unit consistency processing on the data collected in step S101, and output cleaned multi-source data.

[0030] The process involves: removing cloud shadows and detecting and correcting anomalies in remote sensing forest stand data; merging duplicate records and removing measurement anomalies in logging operation and weighbridge measurement data; completing missing links in timber transfer identification data and verifying the consistency between net weight and transfer batches using weighbridge measurement data; correcting anomalies in inventory age data such as negative inventory and age regression; interpolating missing data and standardizing time granularity in road capacity and meteorological risk data; and standardizing the units in processing capacity shift data and energy consumption carbon budget data. The final output is cleaned multi-source data. The source identifier for each record is retained during cleaning for subsequent use in the data reference list in audit voucher data.

[0031] Step S103: Time synchronization of the cleaned multi-source data is performed based on a unified time reference, and time synchronization data is output.

[0032] Using the period as a unified time benchmark, cleaned multi-source data is mapped to the corresponding period. Events spanning multiple periods are handled using a time-proportional allocation method. For example, if a vehicle's weighing occurs near the period boundary, the net weight is allocated to a unique period based on the correspondence between the weighing time and the period. For continuously monitored data such as meteorological risk data, daily-scale data is aggregated into period statistics and written into the time-synchronized data. The aggregation method is summation or averaging within the period, and the aggregation operators used are fixed and consistent with the data type.

[0033] Step S104: Based on the sub-compartment boundaries in the forest compartment ownership data, spatial and node assignments are performed on the time synchronization data, and spatially aligned data is output.

[0034] Step S104 includes: Administrative resource units are determined based on forest compartment ownership data, and remote sensing forest stand data and logging operation measurement data are assigned to the corresponding administrative resource units.

[0035] Industry nodes are determined based on inventory age data, processing capacity shift data, and weighbridge measurement data, and timber transfer identification data is assigned to the corresponding industry nodes.

[0036] Based on road capacity data, the connectivity relationships and capacity parameters of road segments are determined, and the reachability relationships between industry nodes are written into the event alignment dataset.

[0037] Administrative resource units are generated based on the sub-compartment ownership data of forest compartments. Remote sensing forest stand data is spatially overlaid and summarized from pixel to sub-compartment boundary to obtain the period remote sensing statistical value for each administrative resource unit. Logging operation measurement data is then assigned to the corresponding administrative resource unit based on the operation location. Industry node sets are extracted based on inventory age data, processing capacity shift data, and weighbridge measurement data. Timber transfer identification data is assigned to the corresponding industry nodes based on the inbound and outbound node identifiers. The start and end node identifiers from road traffic capacity data are written into the connectivity relationships between industry nodes. Finally, spatially aligned data is output, where each record carries at least a period identifier, an administrative resource unit identifier or an industry node identifier, and the corresponding cleaned fields.

[0038] Step S105: Aggregate the spatially aligned data and organize it by period to output the event aligned dataset.

[0039] Spatially aligned data is organized into event-aligned datasets by period. These datasets use administrative resource units and industry nodes as primary keys and include current remote sensing forest stand statistics, current logging operation measurement values, current timber transfer and weighbridge measurement summaries, current inventory age status, current processing capacity shift status, current demand planning status, current road capacity status, current meteorological risk status, and current energy consumption and carbon budget status. This provides a unique data foundation for subsequent compliance constraint sets, inventory status data, forecast data, and scheduling optimization models.

[0040] Step S2 includes the following sub-steps: Step S201: Analyze the ownership data of forest compartments and extract logging quota parameters, prohibited logging area markers, prohibited logging window parameters, and quarantine risk area markers.

[0041] Based on forest compartment ownership data, logging quota parameters, prohibited logging zone identifiers, and prohibited logging window parameters are extracted, and quarantine risk zone identifiers are solidified according to administrative resource units. For the prohibited logging window parameters, they are standardized into several phase sets, allowing for direct determination of whether logging is permitted in a specific phase.

[0042] Step S202: Analyze the road capacity data and extract the road segment load limit parameters, control window parameters, and maximum capacity parameters.

[0043] Based on road capacity data, segment load limits, maximum capacity parameters, and control window parameters are extracted, and the control window parameters are standardized into several period sets. If a certain period falls within a control window, the maximum capacity parameter of the corresponding segment in that period is set to zero, thus naturally creating an impassable traffic constraint in the scheduling optimization model.

[0044] Step S203: Analyze the energy consumption carbon budget data and extract the processing energy consumption budget parameters and transportation carbon budget parameters.

[0045] Processing energy consumption budget parameters and transportation carbon budget parameters are extracted based on energy consumption carbon budget data. The processing energy consumption budget parameters correspond to the energy consumption caps for processing nodes and periods, while the transportation carbon budget parameters correspond to the transportation carbon emission caps for periods or the carbon emission caps for node pairs. In this embodiment, the budget scope is fixed at the period level.

[0046] Step S204: Organize the logging quota parameters, prohibited logging window parameters, quarantine risk zone identification, road section load limit parameters, control window parameters, processing energy consumption budget parameters, and transportation carbon budget parameters into compliance constraint parameters, and output the compliance constraint set.

[0047] Step S204 includes: The logging quota parameter is converted into a quota constraint that the logging volume does not exceed the quota.

[0048] Convert the prohibited mining zone identifier and prohibited mining window parameters into a window constraint where the logging volume for the prohibited mining period is zero.

[0049] Converting quarantine risk zone identification data into timber transfer identification data must be done through path constraints of designated industry nodes.

[0050] The road segment load limit parameters and maximum traffic capacity parameters are converted into traffic constraints that ensure the transport volume does not exceed the road segment capacity.

[0051] The processing energy consumption budget parameters and transportation carbon budget parameters are converted into budget constraints that ensure that the energy consumption and carbon emissions of each period do not exceed the budget. A rule basis number is generated for each constraint and written into the compliance constraint set.

[0052] The logging quota parameter is instantiated as a quota constraint, the prohibited logging area identifier and prohibited logging window parameter are instantiated as a window constraint, the quarantine risk area identifier is instantiated as a route constraint, the road segment load limit parameter and maximum traffic capacity parameter are instantiated as a traffic constraint, and the processing energy consumption budget parameter and transportation carbon budget parameter are instantiated as a budget constraint. Each instantiated constraint generates a unique rule basis number and writes it into the compliance constraint set. The rule basis number corresponds one-to-one with the constraint type, constraint source field, and its location key in the event alignment dataset, thereby ensuring that subsequent audit voucher data is verifiable.

[0053] Step S3 includes the following sub-steps: Step S301: Based on inventory age data, timber transfer identification data, and weighbridge measurement data, calculate the inbound quantity, outbound quantity, and ending inventory quantity according to industry nodes and periods, and output inventory status data.

[0054] Based on inventory age data, timber circulation identification data, and weighbridge measurement data, the inbound volume, outbound volume, and ending inventory are calculated according to industry nodes and periods. The inbound volume is mainly based on the timber circulation identification data matched with the current inbound node identification, and calibrated using the net weight of the weighbridge measurement data; the outbound volume is mainly based on the timber circulation identification data matched with the current outbound node identification, and is also calibrated using the weighbridge measurement data; the ending inventory is directly read from the current inventory age data at the end of the period, and compared with the inbound and outbound volumes to maintain a balance. If the balance deviation exceeds a preset deviation threshold, the deviation is recorded in the constraint satisfaction record of the audit voucher data.

[0055] Inventory status data The mathematical expression is: ; ; ; in, and Industry node identification, Period identifier in period organization This represents the inventory at the end of the previous period. For the quantity of goods received, For outbound volume, Logging scheduling volume (period) From administrative resource units To the industry node (Dispatch of materials output). For allocation and scheduling volume (period) From the industry node Allocated to industrial nodes (amount) Processing material feeding scheduling quantity (period) At the industry node (the amount of material fed in). For delivery volume (period) From the industry node To demand items Quantity delivered.

[0056] Step S302: Calculate the inventory age index based on the inventory age data, and associate the inventory age index with the inventory status data.

[0057] When the inventory age data already includes the number of days in storage, directly read and verify the consistency of the difference between it and the entry timestamp. If the inventory age data only includes the entry timestamp, calculate the age in days as the difference between the end date of the period and the entry timestamp. Then, associate the inventory age index with the inventory status data according to industry nodes, timber circulation identifiers, and periods to ensure that subsequent quality degradation penalty data can be located to specific inventory batches and specific periods.

[0058] The mathematical expression for the inventory aging index is: ,in, For batch The entry timestamp is mapped to the entry period after the previous entry period.

[0059] Step S303: Based on meteorological risk data, the inventory age index is exposed and corrected to generate quality decay penalty data. The quality decay penalty data characterizes the intensity of loss caused by the increase of inventory age.

[0060] Based on meteorological risk data, inventory aging indicators are subjected to exposure correction to form quality decay penalty data. In this embodiment, a period-specific quality decay intensity is defined for each inventory record. The quality decay intensity is equal to the basic inventory aging decay coefficient multiplied by the meteorological exposure coefficient. The basic inventory aging decay coefficient adopts a piecewise linear approach, setting the low decay interval when the inventory aging days do not exceed a first threshold, the medium decay interval when the inventory aging days are between the first and second thresholds, and the high decay interval when the inventory aging days exceed the second threshold. The decay slope of each interval is given by historical loss statistics or empirical parameters from the competent authority and fixed as model parameters. The meteorological exposure coefficient is calculated from meteorological risk data. The meteorological exposure coefficient is equal to the sum of the continuous rainfall days coefficient multiplied by the continuous rainfall days, plus the extreme temperature coefficient multiplied by the extreme temperature index; the continuous rainfall days coefficient and the extreme temperature coefficient are fixed coefficients. Finally, the quality decay penalty data is calculated by period-specific quality decay intensity multiplied by the ending inventory, written into the event alignment dataset, and used by the objective function of the scheduling optimization model, thereby ensuring that the collected meteorological risk data and inventory aging data are actually used in the optimization.

[0061] The calculation process for the mass degradation penalty data is as follows: First, define the piecewise linear library age function: ; in, For the age of the stock, and The threshold for segmenting storage age is used. , and The slope coefficients of the three linear decay segments; Calculate the meteorological exposure coefficient : ; in, For industrial nodes In the period The meteorological exposure coefficient, For industrial nodes In the period A specific rainfall-related statistic (e.g., number of consecutive rainy days or total rainfall; derived from meteorological risk data). For industrial nodes In the period A certain extreme temperature index (from meteorological risk data). and For exposure coefficient weighting parameters (fixed parameters used to map rainfall and extreme temperatures to exposure coefficients); Calculate the attenuation intensity of each period : ; in, For batch In the period The attenuation intensity, For preset batches In the period The basic decay intensity based on the storage age, For batch At the industry node and period Exposure coefficient; Finally, calculate the mass degradation penalty data. : ; in, For batch In the period The ending inventory.

[0062] Step S304: Generate an inventory baseline based on historical inventory status data and quality decay penalty data.

[0063] An inventory baseline is generated based on historical inventory status data and quality decay penalty data. In this embodiment, the inventory baseline is defined as the median historical ending inventory of each industry node within the same seasonal window, and the corresponding median historical quality decay penalty data. The subsequent comparison logic is as follows: the ending inventory of the current period is compared with the median ending inventory of the inventory baseline to obtain the inventory deviation; the quality decay penalty data of the current period is compared with the median quality decay penalty of the inventory baseline to obtain the decay deviation; the inventory deviation and decay deviation are jointly incorporated into the objective function weight adjustment of the scheduling optimization model to ensure that the model prioritizes reducing nodes with high inventory and high decay.

[0064] The mathematical expression for the inventory baseline is: ; ; in, This is a set of historical periods (used to calculate the historical window of the baseline, such as the most recent periods). For the median operator, For industrial nodes inventory baseline, For industrial nodes The attenuation penalty baseline; The mathematical expressions for inventory deviation and depreciation deviation are: ; ; in, For industrial nodes In the period Inventory deviation For industrial nodes In the period The attenuation deviation.

[0065] Step S4 includes the following sub-steps: Step S401: Calculate the recoverable resources based on remote sensing forest stand data and forest compartment ownership data, and calibrate the data by combining logging operation measurement data to output basic supply data.

[0066] Based on remote sensing forest stand data and forest compartment ownership data, recoverable resources are calculated, and calibrated using logging operation measurement data to output basic supply data. In this embodiment, the estimated recoverable resources for each administrative resource unit are calculated. First, indicators such as canopy height and canopy coverage are extracted from the remote sensing forest stand data and mapped to volume per unit area according to a preset volume inversion relationship. Volume per unit area equals inversion coefficient 1 multiplied by the exponential term of canopy height, plus inversion coefficient 2 multiplied by canopy coverage, plus an intercept term. The inversion coefficients are given by historical sample locations or existing calibration results. Then, the volume per unit area is multiplied by the area of ​​the administrative resource unit to obtain the estimated volume. Subsequently, logging operation measurement data is introduced for calibration. The calibration method is to compare the ratio of the estimated volume to the actual logging operation measurement for the same administrative resource unit in historical periods to obtain the calibration coefficient, and then average it using a sliding window. Finally, the basic supply data equals the estimated volume multiplied by the calibration coefficient and then multiplied by the recoverable ratio parameter set by the competent authority.

[0067] Basic supply data The mathematical expression is: ; ; in, Administrative resource unit In the period Canopy height indices (from remote sensing forest stand data). Administrative resource unit In the period Canopy coverage (from remote sensing forest stand data). Administrative resource unit In the period The estimated volume per unit area (inversion result). The area represents the administrative resource unit (derived from forest compartment ownership data). , and These are the inversion coefficients. These are the acceptable proportion parameters (fixed parameters, administrative management standards). It is a power exponent (fixed parameter).

[0068] The basic supply quantity after calibration is: ; in, This is a set of sliding windows (several historical periods) used for calibration. Administrative resource unit In historical periods The logging operation measurement values ​​(from logging operation measurement data).

[0069] Step S402: Based on meteorological risk data and road capacity data, accessibility and risk reduction are applied to the basic supply data to output recoverable supply forecast data.

[0070] Based on meteorological risk data and road capacity data, accessibility and risk reductions are applied to the basic supply data to output recoverable supply forecast data. Accessibility reduction is derived from road capacity data. If a control window exists or the maximum capacity parameter is zero for the accessible road segment from an administrative resource unit to the nearest industrial node in a given period, the accessibility reduction coefficient for that period is set to zero; otherwise, the accessibility reduction coefficient equals the ratio of the minimum to maximum capacity parameter of that path to the baseline capacity, limited to between zero and one. Risk reduction is derived from meteorological risk data. The risk reduction coefficient equals one minus the risk level mapping coefficient multiplied by the risk level, limited to between zero and one. The final recoverable supply forecast data equals the basic supply data multiplied by the accessibility reduction coefficient and then multiplied by the risk reduction coefficient. It is still subject to quota and window constraints imposed by the compliance constraint set to ensure that the forecast data does not generate unenforceable supply boundaries.

[0071] The mathematical expression for the available supply forecast data is: ; in, For road section In the period Maximum capacity parameters (from road capacity data). For road section The baseline traffic capacity (fixed baseline or historical normal value). Risk mapping coefficient (fixed parameter). It is classified as a risk level.

[0072] Step S403: Generate administrative demand forecast data based on demand planning data, and calculate the demand urgency index. The demand urgency index is jointly determined by the delivery window, supply guarantee level, and default impact item.

[0073] Administrative demand forecasting data is generated based on demand planning data. In this embodiment, administrative demand forecasting data primarily uses demand planning data, and historical execution feedback data is incorporated to adjust the reliability of the demand plan. The reliability adjustment method is as follows: calculate the plan fulfillment rate of the same demand subject in historical periods. The plan fulfillment rate equals the historical actual delivery volume divided by the historical planned demand volume, and the reliability coefficient is obtained by weighting the nearest periods. The administrative demand forecasting data equals the demand volume of the demand planning data multiplied by the reliability coefficient, plus the supply guarantee compensation parameter specified by the competent department. In this way, both demand planning data and execution feedback data are used for demand forecasting, and the forecasting caliber remains consistent with administrative management.

[0074] Administrative demand forecast data The mathematical expression is: ; Among them, historical periods For demand items The actual delivery volume (from execution feedback data, which can be obtained by summarizing weighbridge measurement data and timber transfer identification data). For historical periods For demand items The demand, For the supply compensation amount parameter (fixed parameter, used for administrative caliber correction). This refers to the demand quantity in the demand planning data (from the demand planning data).

[0075] Demand Urgency Index The mathematical expression is: ; in, , and The weighting parameters for the urgency index, The supply guarantee level score is obtained by mapping the supply guarantee level from the demand planning data. This is the shortest window baseline constant (a fixed parameter used to quantify the urgency of a shorter window). The penalty for breach of contract (fixed or given by the rules of the competent authority). This refers to the remaining delivery window length or the remaining number of periods (calculated from the delivery window in the demand planning data).

[0076] Step S404: Write the demand urgency index into the administrative demand forecast data to characterize the supply guarantee priority.

[0077] The urgency index is incorporated into administrative demand forecasting data to characterize supply priority. The urgency index is jointly determined by the delivery window, supply level, and default impact factor. In this embodiment, the urgency index equals the supply level weight multiplied by the supply level score, plus the delivery window weight multiplied by the delivery window urgency score, plus the default impact weight multiplied by the default impact score. The delivery window urgency score is equal to the preset minimum window days divided by the remaining window days with an upper limit. The default impact score is given and fixed by the competent authority's administrative impact assessment of the demand. The urgency index is used as the coefficient for delay penalties in the scheduling optimization model, ensuring that demands with high supply levels and urgent windows are prioritized for satisfaction in the optimal solution.

[0078] Step S5 includes the following sub-steps: Step S501: Establish a scheduling optimization model and set the logging scheduling quantity, transportation scheduling quantity, allocation scheduling quantity and processing material input scheduling quantity as decision objects.

[0079] Within a given period, the management system establishes a scheduling optimization model and defines decision-making objects. These objects include logging scheduling volume, road segment transportation volume, allocation scheduling volume, processing and material input scheduling volume, and delivery volume. The logging scheduling volume is located by administrative resource unit identifiers, industry node identifiers, and period identifiers; the road segment transportation volume is located by road segment identifiers and period identifiers; the allocation scheduling volume is located by outgoing industry node identifiers, incoming industry node identifiers, and period identifiers; the processing and material input scheduling volume is located by processing node identifiers and period identifiers; and the delivery volume is located by industry node identifiers, demand item identifiers, and period identifiers. This ensures that the administrative resource units, industry nodes, road segments, and demand items in the event alignment dataset can all correspond one-to-one with the decision-making objects.

[0080] In the process of establishing the scheduling optimization model, we first define the unmet needs. : ; in, Forecast data for administrative needs.

[0081] The objective function is then: ; in, This refers to the segment transport volume (the expression of the transport plan at the segment level; used to consume the maximum capacity parameter, load limit parameter, and control window parameter). This is the transportation cost coefficient for the road segment (a fixed parameter). This is the processing cost coefficient (fixed parameter). , , , and These are the weights of each term in the objective function.

[0082] The quota constraints (logging limit parameters) of the scheduling optimization model are: ; in, Administrative resource unit In the period The amount of logging and dispatching, This is the logging quota parameter (to the right of the quota constraint, derived from the forest compartment ownership data analysis).

[0083] The window constraint (no-sampling window parameter, indicated by an indicator in the example) of the scheduling optimization model is as follows: ; in, For the prohibited sampling window quantity (if the period) The value is 1 if the sample is within the restricted sampling window, and 0 otherwise (from the restricted sampling window parameter mapping). It is a large number constant (used to express a technical constant that is 0 when mining is prohibited as a linear form).

[0084] The road traffic constraints (maximum capacity parameter) of the scheduling optimization model are: ; The load constraints (segment load parameters) of the scheduling optimization model are: ; in, These are the load limit parameters for road sections (derived from road capacity data).

[0085] The processing capacity constraints (processing capacity shift data) of the scheduling optimization model are as follows: ; in, The number of available shifts (from processing capacity shift data). Capacity per shift (from processing capacity shift data).

[0086] The energy consumption budget constraint (processing energy consumption budget parameter) of the scheduling optimization model is: ; in, Energy consumption per unit of material input or energy consumption per unit of output (derived from processing capacity shift data or fixed parameters). For processing energy consumption budget parameters (from energy consumption carbon budget data).

[0087] The transportation carbon budget constraint (transportation carbon budget parameters) for the scheduling optimization model is as follows: ; in, Carbon emission coefficient for transportation per unit of road segment (fixed parameter or administrative parameter). For transportation carbon budget parameters (from energy consumption carbon budget data).

[0088] Planned and returned deviation rate for: ; in, This refers to a specific type of planned quantity (derived from scheduling scheme data). This refers to a specific type of execution feedback data (derived from execution feedback data, which is a summary of timber transfer identification data, weighbridge measurement data, and logging operation measurement data). It is a very small positive number (to prevent numerical stability constants where the denominator is 0).

[0089] The rolling update trigger condition for the scheduling optimization model is: ; ; in, The preset deviation threshold and the preset risk level threshold are defined.

[0090] Step S502 sets the optimization objectives as minimizing ending inventory, minimizing losses corresponding to quality degradation penalty data, minimizing delay penalties for administrative demand forecast data, and minimizing transportation and processing costs.

[0091] The management system establishes inventory conservation constraints based on inventory status data and uses them as the basic constraint input for the scheduling optimization model. This ensures that the ending inventory of each industry node in each period satisfies the conservation relationship with the inbound, outbound, and processing material scheduling quantities. The inbound quantity consists of the inflow of logging scheduling quantity and the inflow of allocation scheduling quantity, while the outbound quantity consists of the outflow of allocation scheduling quantity and the delivery quantity. The consistency between the inbound and outbound quantities is verified using inventory age data and weighbridge measurement data. The verification deviation is recorded in the constraint satisfaction record of the audit voucher data for review.

[0092] In step S503, the quota constraints, window constraints, path constraints, access constraints, and budget constraints in the compliance constraint set are parsed into constraint terms, and together with the inventory conservation constraints corresponding to the inventory status data, they are used as constraint inputs for the scheduling optimization model.

[0093] The inventory baseline and quality decay penalty data are mapped to inventory penalty parameters and decay penalty parameters, and the available supply forecast data and administrative demand forecast data are mapped to supply boundary parameters and demand boundary parameters, which are used as the target parameters and boundary parameters input to the scheduling optimization model.

[0094] The management end parses each compliance constraint into directly calculable constraint items and inputs them into the scheduling optimization model. Specifically, for the same administrative resource unit, the sum of logging scheduling quantities within a statistical window is limited to the logging quota parameter to form a quota constraint; for periods covered by the prohibited logging window parameter, the logging scheduling quantity is limited to zero to form a window constraint; for the same road segment and period, the road segment transportation volume is limited to the maximum traffic capacity parameter and the road segment load limit parameter to form a traffic constraint; for the same processing node and period, the energy consumption corresponding to the processing input scheduling quantity is limited to the processing energy consumption budget parameter, and the carbon emission corresponding to the road segment transportation volume in the same period is limited to the transportation carbon budget parameter to form a budget constraint; simultaneously, an inventory conservation constraint is established based on inventory status data. The management end converts the inventory baseline and quality decay penalty data into inventory penalty parameters and decay penalty parameters and writes them into the optimization objective; the recoverable supply forecast data is converted into the upper limit parameter of the logging scheduling quantity supply; and the administrative demand forecast data is converted into the demand boundary parameter of the delivery quantity. This allows the compliance constraint set, inventory baseline, and forecast data to enter the same model in a fixed manner and to be solved reproducibly.

[0095] Step S504: Solve the scheduling optimization model and output the scheduling scheme data, which includes phased logging plans, transportation plans, allocation plans, processing plans and delivery plans.

[0096] Step S504 includes: When the scheduling optimization model satisfies that all decision objects are continuously allocable quantities, the scheduling optimization model is set as a linear programming model and the linear programming solution strategy is used to output the scheduling scheme data.

[0097] When there are discrete execution constraints for the entire shift or the entire vehicle in the processing capacity shift data or road traffic capacity data, or when there are minimum operation batch constraints for the logging scheduling quantity corresponding to the logging operation measurement data, the scheduling optimization model is set as a mixed integer linear programming model and the branch and bound solution strategy is used to output the scheduling scheme data.

[0098] If the compliance constraint set makes the model infeasible, the hard constraints corresponding to the rule basis number remain unchanged, and the administrative demand forecast data with low demand urgency index is subject to controllable degradation, and the scheduling scheme data containing degradation records is output.

[0099] The management end sets the model type of the scheduling optimization model based on the discrete execution constraint parameters. If there are discrete execution constraint parameters for the entire shift in the processing capacity shift data, or discrete execution constraint parameters for the entire vehicle in the road traffic capacity data, or minimum operation batch constraint parameters for the logging scheduling quantity corresponding to the logging operation measurement data, then the scheduling optimization model is set to a mixed integer linear programming model. Otherwise, the scheduling optimization model is set to a linear programming model. After the model type is set, the management end solves the scheduling optimization model and outputs the scheduling scheme data. The model type, the source field of the discrete execution constraint parameters, and the enabled compliance constraint set rule number are written into the audit voucher data for review.

[0100] Step S6 includes the following sub-steps: Step S601: Generate logging execution instruction data, transportation execution instruction data, allocation execution instruction data and processing scheduling execution instruction data based on the scheduling scheme data, and send them to the management terminal.

[0101] Based on the scheduling plan data, logging execution instructions, transportation execution instructions, allocation execution instructions, and processing scheduling execution instructions are generated and distributed to the management terminal. The logging execution instructions include administrative resource units, periods, logging volumes, and corresponding receiving industry nodes; the transportation execution instructions include starting and ending nodes, periods, and transport volumes, along with road segment identifiers linked to road capacity data for verification; the allocation execution instructions include sending and receiving nodes, periods, and allocation volumes, along with batch generation rules linked to timber transfer identifier data; and the processing scheduling execution instructions include processing nodes, periods, processing material allocation volumes, and shift allocation results for corresponding processing capacity shifts.

[0102] Step S602: Collect execution feedback data based on timber transfer identification data, weighbridge measurement data and logging operation measurement data, write the execution feedback data into the event alignment dataset and update the inventory status data and inventory age data.

[0103] Based on timber transfer identification data, weighbridge measurement data, and logging operation measurement data, execution feedback data is collected and written into an event alignment dataset to update inventory status and inventory age data. During the write-back process, timber transfer identification data is used as the unique link primary key. Weighbridge measurement data is used to verify the execution volume differences of transportation execution instructions, and logging operation measurement data is used to verify the execution volume differences of logging execution instructions. The difference is written into the event alignment dataset as an execution deviation for subsequent deviation threshold judgment and rolling updates.

[0104] Step S603: Generate audit voucher data, which includes a data reference list, a rule basis number list, constraint satisfaction records, and inventory reduction comparison records.

[0105] Generate audit voucher data, which includes at least a data reference list, a rule basis number list, constraint satisfaction records, and inventory reduction comparison records. The data reference list records the event-aligned dataset fields and time ranges actually invoked in the current solution; the rule basis number list records the activated constraints in the compliant constraint set and their basis numbers; the constraint satisfaction record records the calculated left-hand side value and right-hand side threshold of each hard constraint and indicates whether it is satisfied; the inventory reduction comparison record records the changes in inventory deviation and decay deviation relative to the inventory baseline before and after the execution of the scheduling plan data, used for administrative performance evaluation.

[0106] Step S604: When the deviation of the returned data from the scheduling scheme data exceeds the preset deviation threshold or the meteorological risk data triggers the preset risk condition, a rolling update is triggered and steps S2 to S6 are re-executed.

[0107] When the deviation of the executed feedback data from the scheduling plan data exceeds a preset deviation threshold, or when meteorological risk data triggers a preset risk condition, a rolling update is triggered, and steps S2 to S6 are re-executed. The judgment logic for the deviation exceeding the preset deviation threshold is as follows: the logging execution deviation, transportation execution deviation, allocation execution deviation, and processing execution deviation are compared with their planned quantities. If the absolute value of any deviation to the planned quantity exceeds the preset deviation threshold, the process is triggered. The risk condition trigger logic is as follows: if the risk level of the meteorological risk data exceeds the preset risk level threshold and the corresponding period is near the boundary of the prohibited harvesting window parameter, or if the road capacity data shows a new control window parameter, the process is triggered. After triggering, the event alignment dataset is updated, the compliance constraint set is re-instantiated, and the inventory status data and quality decay penalty data are recalculated. Then, the scheduling optimization model is solved to output new scheduling plan data, thereby achieving closed-loop and interpretable updates for administrative scheduling.

[0108] This invention constructs a unified event-aligned dataset by collecting remote sensing forest stand data, forest compartment ownership data, logging operation measurement data, timber transfer identification data, weighbridge measurement data, inventory age data, processing capacity shift data, demand planning data, road capacity data, meteorological risk data, and energy consumption carbon budget data. It also instantiates logging quotas, logging ban windows, quarantine routes, road section load limits and control windows, processing energy consumption budgets, and transportation carbon budgets as compliance constraint sets, making the scheduling process calculable and verifiable under administrative rules. Based on this, it calculates inventory status data and quality degradation penalty data and establishes an inventory baseline, explicitly incorporating the quality degradation losses caused by inventory quantity and age into the optimization objective. This prompts the scheme to prioritize reducing the backlog at high-age and high-risk nodes, achieving inventory reduction and quality loss control. The system synchronizes supply and demand forecasts; it further generates available supply forecasts and administrative demand forecasts, and quantifies demand urgency indicators to prioritize key projects and livelihood supply needs within the model, while avoiding excessive pressure on general needs; it solves the problem using linear programming or mixed-integer linear programming to output integrated scheduling data for logging, transportation, allocation, processing, and delivery plans, and generates audit voucher data including data reference lists, rule basis number lists, and constraint satisfaction records to improve the traceability of administrative decisions; finally, it generates execution feedback data based on timber transfer identification and weighbridge measurement, triggering rolling updates according to deviation thresholds and risk conditions to achieve rapid rescheduling in emergencies such as road interruptions and sudden increases in weather risks, thereby improving the real-time performance, compliance, and stable supply guarantee capabilities of the scheduling.

[0109] Those skilled in the art will understand that embodiments of the present invention can be provided as methods, systems, or computer program products. Therefore, the present invention can take the form of a completely hardware embodiment, a completely software embodiment, or an embodiment combining software and hardware aspects. Furthermore, the present invention can take the form of a computer program product implemented on one or more computer-usable storage media containing computer-usable program code. The storage medium can be implemented by any type of volatile or non-volatile storage device or a combination thereof, such as Static Random Access Memory (SRAM), Electrically Erasable Programmable Read-Only Memory (EEPROM), Erasable Programmable Read Only Memory (EPROM), Programmable Read-Only Memory (PROM), Read-Only Memory (ROM), magnetic storage, flash memory, magnetic disk, or optical disk. These computer program instructions may also be stored in a computer-readable storage medium that can direct a computer or other programmable data processing device to function in a particular manner, such that the instructions stored in the computer-readable storage medium produce an article of manufacture including instruction means, which are implemented in a process Figure 1 One or more processes and / or boxes Figure 1 The function specified in one or more boxes.

[0110] It should be noted that the above embodiments are only used to illustrate the technical solutions of the present invention and are not intended to limit it. Although the present invention has been described in detail with reference to preferred embodiments, those skilled in the art should understand that modifications or equivalent substitutions can be made to the technical solutions of the present invention without departing from the spirit and scope of the technical solutions of the present invention, and all such modifications or substitutions should be covered within the protection scope of the present invention.

Claims

1. A forestry industry scheduling optimization method based on multi-source data, characterized in that, Includes the following steps: Step S1: Collect remote sensing forest stand data, forest compartment ownership data, logging operation measurement data, timber transfer identification data, weighbridge measurement data, inventory age data, processing capacity shift data, demand planning data, road traffic capacity data, meteorological risk data, and energy consumption carbon budget data; clean and spatiotemporally align the data to generate an event-aligned dataset. Step S2: Generate a set of compliance constraints based on the event alignment dataset; Step S3: Calculate inventory status data and quality decay penalty data, and establish an inventory baseline; Step S4: Generate available supply forecast data and administrative demand forecast data; Step S5: Compile the compliance constraint set into constraint matrix data, compile the inventory baseline and forecast data into target coefficient data and boundary vector data, construct the scheduling optimization matrix model and solve for the output scheduling scheme data; Step S6: Generate execution instruction data and audit voucher data based on the scheduling scheme data and update them on a rolling basis; Step S3 includes the following sub-steps: Step S301: Based on inventory age data, timber transfer identification data and weighbridge measurement data, calculate the inbound quantity, outbound quantity and ending inventory quantity according to industry nodes and periods, and output inventory status data. Step S302: Calculate the inventory age index based on the inventory age data, and associate the inventory age index with the inventory status data; Step S303: Based on meteorological risk data, the inventory age index is exposed and corrected to generate quality decay penalty data, which characterizes the intensity of loss caused by the increase of inventory age. Step S304: Generate an inventory baseline based on historical inventory status data and quality decay penalty data. Step S5 includes the following sub-steps: Step S501: Establish a scheduling optimization model and set the logging scheduling quantity, transportation scheduling quantity, allocation scheduling quantity and processing and feeding scheduling quantity as decision objects; Step S502: The optimization objectives are to minimize the ending inventory, minimize the loss corresponding to the quality degradation penalty data, minimize the delay penalty of the administrative demand forecast data, and minimize the transportation and processing costs. Step S503: The quota constraints, window constraints, path constraints, access constraints and budget constraints in the compliance constraint set are parsed into constraint terms, and together with the inventory conservation constraints corresponding to the inventory status data, they are used as constraint inputs for the scheduling optimization model. The inventory baseline and quality decay penalty data are mapped to inventory penalty parameters and decay penalty parameters, and the available supply forecast data and administrative demand forecast data are mapped to supply boundary parameters and demand boundary parameters, which are used as the target parameters and boundary parameters input of the scheduling optimization model. Step S504: Solve the scheduling optimization model and output the scheduling scheme data, which includes phased logging plans, transportation plans, allocation plans, processing plans and delivery plans; Step S6 includes the following sub-steps: Step S601: Generate logging execution instruction data, transportation execution instruction data, allocation execution instruction data, and processing scheduling execution instruction data based on the scheduling scheme data, and send them to the management terminal; Step S602: Collect execution feedback data based on timber transfer identification data, weighbridge measurement data and logging operation measurement data, write the execution feedback data into the event alignment dataset and update the inventory status data and inventory age data. Step S603: Generate audit voucher data, which includes a data reference list, a rule basis number list, constraint satisfaction records, and inventory reduction comparison records. Step S604: When the deviation of the returned data from the scheduling scheme data exceeds the preset deviation threshold or the meteorological risk data triggers the preset risk condition, a rolling update is triggered and steps S2 to S6 are re-executed.

2. The forestry industry scheduling optimization method based on multi-source data as described in claim 1, characterized in that, Step S1 includes the following sub-steps: Step S101: Obtain remote sensing forest stand data, forest compartment ownership data, logging operation measurement data, timber transfer identification data, weighbridge measurement data, inventory age data, processing capacity shift data, demand planning data, road traffic capacity data, meteorological risk data, and energy consumption carbon budget data. Step S102: Perform missing data removal, outlier correction and unit consistency processing on the data collected in step S101, and output cleaned multi-source data. Step S103: Perform time synchronization on the cleaned multi-source data based on a unified time reference, and output time synchronization data; Step S104: Based on the sub-compartment boundaries in the forest compartment ownership data, spatial and node assignments are performed on the time synchronization data, and spatially aligned data is output. Step S105: Aggregate the spatially aligned data and organize it by period to output the event aligned dataset.

3. The forestry industry scheduling optimization method based on multi-source data as described in claim 2, characterized in that, Step S104 includes: Administrative resource units are determined based on forest compartment ownership data, and remote sensing forest stand data and logging operation measurement data are assigned to the corresponding administrative resource units. Based on inventory age data, processing capacity shift data, and weighbridge measurement data, industry nodes are determined, and timber transfer identification data is assigned to the corresponding industry nodes. Based on road capacity data, the connectivity relationships and capacity parameters of road segments are determined, and the reachability relationships between industry nodes are written into the event alignment dataset.

4. The forestry industry scheduling optimization method based on multi-source data as described in claim 1, characterized in that, Step S2 includes the following sub-steps: Step S201: Analyze the ownership data of forest compartments and extract logging quota parameters, prohibited logging area markers, prohibited logging window parameters, and quarantine risk area markers; Step S202: Analyze the road capacity data and extract the road segment load limit parameters, control window parameters, and maximum capacity parameters; Step S203: Analyze the energy consumption carbon budget data and extract the processing energy consumption budget parameters and transportation carbon budget parameters; Step S204: Organize the logging quota parameters, prohibited logging window parameters, quarantine risk zone identification, road section load limit parameters, control window parameters, processing energy consumption budget parameters, and transportation carbon budget parameters into compliance constraint parameters, and output the compliance constraint set; Step S204 includes: Convert the logging quota parameter into a quota constraint that the logging volume does not exceed the quota; Convert the prohibited mining zone identifier and prohibited mining window parameters into a window constraint where the logging volume for the prohibited mining period is zero; Converting quarantine risk zone markers into timber transfer marker data requires path constraints through designated industry nodes; The road segment load limit parameters and maximum traffic capacity parameters are converted into traffic constraints that ensure the transport volume does not exceed the road segment capacity. The processing energy consumption budget parameters and transportation carbon budget parameters are converted into budget constraints that ensure that the energy consumption and carbon emissions of each period do not exceed the budget. A rule basis number is generated for each constraint and written into the compliance constraint set.

5. The forestry industry scheduling optimization method based on multi-source data as described in claim 1, characterized in that, Step S4 includes the following sub-steps: Step S401: Calculate the recoverable resources based on remote sensing forest stand data and forest compartment ownership data, and calibrate them in conjunction with logging operation measurement data to output basic supply data. Step S402: Based on meteorological risk data and road capacity data, accessibility and risk reduction are performed on the basic supply data, and the available supply forecast data is output. Step S403: Generate administrative demand forecast data based on demand planning data, and calculate the demand urgency index. The demand urgency index is jointly determined by the delivery window, supply guarantee level, and default impact item. Step S404: Write the demand urgency index into the administrative demand forecast data to characterize the supply guarantee priority.

6. The forestry industry scheduling optimization method based on multi-source data as described in claim 1, characterized in that, Step S504 includes: When the scheduling optimization model satisfies that all decision objects are continuously allocable quantities, the scheduling optimization model is set as a linear programming model and the linear programming solution strategy is used to output the scheduling scheme data. When there are discrete execution constraint parameters for the entire shift or the entire vehicle in the processing capacity shift data or road traffic capacity data, or when there are minimum operation batch constraint parameters for the logging scheduling quantity corresponding to the logging operation measurement data, the scheduling optimization model is set as a mixed integer linear programming model and the branch and bound solution strategy is used to output the scheduling scheme data. If the compliance constraint set makes the model infeasible, the hard constraints corresponding to the rule basis number remain unchanged, and the administrative demand forecast data with low demand urgency index is subject to controllable degradation, and the scheduling scheme data containing degradation records is output.

Images