Intelligent period and current fusion sales execution system and method

Abstract

The invention relates to a sales execution system and method based on intelligent period and current fusion, and the method comprises the steps: the system receives a customer order or automatically generates a sales plan, carries out the period and current analysis when an inventory monitoring module inquires that there is enough inventory meeting the demand of the order, and judges whether the price difference rate exceeds a threshold value delta or not; if so, marking as preferential futures sales; the AI decision engine starts multi-objective optimization calculation to obtain a comprehensive income index K, if Kgt; if the threshold value is theta, selecting a futures sales + spot replenishment strategy, and if Klt; if-theta, selecting a pure spot sales strategy, otherwise, selecting a hybrid strategy; and performing risk check on the generated sales scheme, executing the sales instruction after the risk check is completely passed, monitoring the order execution state in real time, and tracking the spot delivery progress, the futures transaction condition and the customer payment state. According to the method, multiple dimensions such as income, inventory and risk are comprehensively considered, and the problem of local optimization possibly caused by single-target optimization is avoided.

Description

Technical Field

[0001] This invention relates to the field of big data processing, and in particular to an intelligent futures-spot fusion sales execution system and method. Background Technology

[0002] In the commodity trading sector, companies need to sell physical products through the spot market and also utilize the futures market for risk management and hedging. In traditional sales models, companies typically treat spot sales and futures operations as two separate business processes, managed by different departments. The spot sales department focuses on daily operations such as customer orders, delivery plans, and payment collection, while the futures department concentrates on establishing hedging positions and managing portfolio risk. With increasing market competition and frequent price fluctuations, companies increasingly need an AI-powered sales system that can comprehensively consider both the spot and futures markets. Such a system must not only handle traditional order management functions but also AI-drivenly formulate sales strategies based on factors such as spot-futures price relationships, inventory levels, and market expectations, selecting the optimal sales channels and timing to maximize corporate profits.

[0003] However, the market currently lacks a truly integrated futures and spot sales system. Existing ERP systems primarily handle spot sales processes, while futures trading software focuses on trade execution, resulting in a lack of effective data interaction and strategy synergy between the two. Therefore, companies currently face the following main problems in commodity sales: 1. Lack of AI support for sales decisions: Sales personnel rely mainly on experience to determine when to sell through spot channels and when to hedge through the futures market. The lack of quantitative analysis tools makes it easy to miss optimal sales opportunities or choose suboptimal sales channels, leading to unsatisfactory sales returns; 2. Disconnect between futures hedging and spot sales: After signing spot contracts, the sales department needs to manually notify the futures department to establish corresponding hedging positions. Information transmission is delayed and prone to errors, potentially leading to untimely hedging or mismatched hedging quantities, exposing the company to price risk; 3. Inability to monitor futures-spot price differences in real time: Companies struggle to promptly identify arbitrage opportunities or avoid unfavorable price relationships. When futures prices are significantly higher than spot prices, priority should be given to using the futures market. 4. Sales and replenishment from the spot market are not automatically recognized and executed by the existing system; 5. Inventory management and sales strategies lack linkage, and sales decisions fail to fully consider factors such as inventory costs, warehousing capacity, and inventory turnover, which may lead to inventory backlog or shortage, affecting capital efficiency and customer satisfaction; 6. The sales execution process involves a lot of manual operation. From receiving orders, confirming inventory, arranging delivery to placing hedging orders, each step requires manual intervention, resulting in low operational efficiency and easy errors due to human negligence, increasing operating costs and risks; 7. There is a lack of risk control mechanisms. Problems such as customer credit risk, large price fluctuation risk, and excessive hedging positions may occur during the sales process. The existing system cannot provide real-time risk monitoring and automatic early warning functions. Once a risk event occurs, the company is often caught off guard. Summary of the Invention

[0004] The purpose of this invention is to overcome the shortcomings of the prior art and provide an intelligent futures and spot integrated sales execution system and method, which solves the deficiencies of the prior art.

[0005] The objective of this invention is achieved through the following technical solution: an intelligent futures-spot integrated sales execution system, comprising an order management module, an inventory monitoring module, a futures-spot analysis module, an AI decision engine, a risk assessment module, and an automatic execution module; The order management module is configured to receive and process customer information. The inventory monitoring module is configured to track the enterprise's inventory status in real time. The futures-spot analysis module is configured to collect futures market data and spot market quotes in real time, calculate the futures-spot price difference, basis, and arbitrage opportunities, and analyze market trends and price movements. The AI ​​decision engine is configured to generate the optimal sales plan based on information such as order demand, inventory status, futures and spot price relationships, and market expectations, using optimization algorithms. The risk assessment module is configured to perform a multi-dimensional risk assessment of the sales plan. The automatic execution module is configured to automatically execute spot delivery instructions, futures trading instructions, and inventory transfer instructions based on the sales plan generated by the decision engine, and monitor the execution status in real time.

[0006] The system also includes a customer management module and a data center; The customer management module is configured to maintain a customer information database, including customer credit registration, historical transaction records, payment habits, and preferred products, providing customer profiles for sales decisions. The data center is configured to use a combination of time-series and relational databases to store historical market data, order data, inventory data, and transaction records to support big data analytics and machine learning model training.

[0007] The method of using optimization algorithms to generate the optimal sales plan includes: Obtain the order quantity Q, current inventory I, total inventory S, and expected demand D; Calculate the inventory adequacy R = (IS) / D. If R > 1.5, it indicates that the inventory is sufficient and you tend to sell quickly to reduce inventory. If R is less than 0.8, it indicates that the inventory is tight and you tend to delay sales or raise prices. Calculate the futures sales revenue E_futures = Q × P_futures - Q × (transaction costs + delivery costs); Calculate the spot sales revenue Espot = Q × Pspot - Q × (logistics cost + warehousing cost); The comprehensive return index K is calculated as follows: K = α × (E futures - E spot) + β × R - γ × risk coefficient, where α, β, and γ are weighting parameters. If K > threshold θ, then choose the strategy of futures sales + spot replenishment with futures sales priority; if K < -θ, then choose the pure spot sales strategy; otherwise, choose the mixed strategy.

[0008] The multi-dimensional risk assessment includes: Customer credit check: Check the customer's credit record. If the requirements are not met, a prepayment or guarantee is required. Price volatility risk: Calculate the historical price volatility σ. If σ > a set threshold, reduce the proportion of futures sales or increase hedging efforts. Inventory risk: After the price difference is sold, is the inventory below the safety stock? If so, an emergency procurement plan will be triggered. Hedging ratio check: Calculate the proportion of hedging positions to total sales volume and ensure it is within the set range; Funding risk: Check whether the futures margin required exceeds the safe ratio of available funds.

[0009] A method based on an intelligent futures-spot fusion sales execution system, the method comprising: S1. When the system receives customer orders or automatically generates sales plans, it performs spot-futures analysis when the inventory monitoring module finds that there is enough inventory to meet the order demand, and determines whether the price difference rate exceeds the threshold δ. If it does, it is marked as a priority futures sale. S2. The AI ​​decision engine initiates multi-objective optimization calculations to obtain the comprehensive return index K. If K > threshold θ, then the futures sales + spot replenishment strategy with futures sales priority is selected. If K < -θ, then the pure spot sales strategy is selected. Otherwise, the mixed strategy is selected. S3. Conduct risk checks on the generated sales plan. Once all risk checks are passed, execute the sales order and monitor the order execution status in real time, tracking the progress of spot delivery, futures transactions, and customer payment status.

[0010] The method further includes: S4. Once the spot goods are delivered or the futures position is closed, the profit calculation is performed automatically, the actual profit is compared with the expected profit, and the reasons for the deviation are analyzed. S5. Record the complete sales data to the data center and determine whether to continue monitoring the market and processing new orders.

[0011] The AI ​​decision engine initiates multi-objective optimization calculations to obtain the comprehensive return index K, which includes: Obtain the order quantity Q, current inventory I, total inventory S, and expected demand D; Calculate the inventory adequacy R = (IS) / D. If R > 1.5, it indicates that the inventory is sufficient and you tend to sell quickly to reduce inventory. If R is less than 0.8, it indicates that the inventory is tight and you tend to delay sales or raise prices. Calculate the futures sales revenue E_futures = Q × P_futures - Q × (transaction costs + delivery costs); Calculate the spot sales revenue Espot = Q × Pspot - Q × (logistics cost + warehousing cost); The comprehensive return index K is calculated as follows: K = α × (E futures - E spot) + β × R - γ × risk coefficient, where α, β, and γ are weighting parameters.

[0012] The risk assessment includes: Customer credit check: Check the customer's credit record. If the requirements are not met, a prepayment or guarantee is required. Price volatility risk: Calculate the historical price volatility σ. If σ > a set threshold, reduce the proportion of futures sales or increase hedging efforts. Inventory risk: After the price difference is sold, is the inventory below the safety stock? If so, an emergency procurement plan will be triggered. Hedging ratio check: Calculate the proportion of hedging positions to total sales volume and ensure it is within the set range; Funding risk: Check whether the futures margin required exceeds the safe ratio of available funds.

[0013] The execution of the sales instruction includes: If spot sales are selected, an outbound instruction is sent to the WMS system to generate a delivery note and notify the logistics department to arrange delivery. If you choose to sell futures, send a sell order to the futures trading system to establish a short hedging position. If a mixed strategy is adopted, spot and futures operations will be executed in proportion to each other. Update the inventory system and lock the corresponding inventory to prevent duplicate sales.

[0014] This invention offers the following advantages: an intelligent futures-spot integrated sales execution system and method capable of real-time analysis of futures-spot price relationships, automatic identification of arbitrage opportunities, and AI-driven selection of the optimal sales channel. Compared to traditional manual decision-making, sales revenue increases by an average of 4-6%, significantly improving corporate profitability. Upon receiving a spot order, the system automatically assesses whether a futures hedging position needs to be established and automatically executes the hedging operation, improving hedging timeliness from hours to seconds, effectively reducing price volatility risk, and increasing hedging coverage to over 95%. By comprehensively considering multiple dimensions such as revenue, inventory, and risk, it avoids the local optima problem that may arise from optimizing a single objective. Actual testing showed that the system-generated solutions outperformed manual solutions by approximately 15-20% in overall performance. It dynamically adjusts sales strategies based on real-time inventory levels, accelerating inventory reduction when inventory is plentiful and delaying sales or raising prices when inventory is tight, increasing inventory turnover by over 30% and reducing capital tied up in inventory and warehousing costs. It achieves end-to-end automation of the sales process, from order receipt, decision generation, risk assessment to execution, requiring no manual intervention and improving processing efficiency by over 80%. The error rate of human operation is reduced to near zero, significantly reducing operating costs and operational risks. The system provides real-time monitoring and early warning of customer credit risk, price risk, inventory risk, and financial risk. Each sales transaction must pass a risk check before execution; if a risk exceeds the limit, operation is immediately suspended and management is notified. No major risk events have occurred since the system's operation began. Attached Figure Description

[0015] Figure 1 This is a schematic diagram of the system of the present invention; Figure 2This is a schematic flowchart of the method of the present invention; Figure 3 A flowchart illustrating the decision-making logic of an AI decision engine; Figure 4 This is a flowchart of the risk verification process for the risk assessment module. Figure 5 A flowchart of a multi-objective optimization algorithm; Figure 6 This is a flowchart illustrating the training and application process of the machine learning module. Detailed Implementation

[0016] To make the objectives, technical solutions, and advantages of the embodiments of this application clearer, the technical solutions of the embodiments of this application will be clearly and completely described below with reference to the accompanying drawings. Obviously, the described embodiments are only a part of the embodiments of this application, and not all of the embodiments. The components of the embodiments of this application described and shown in the accompanying drawings can generally be arranged and designed in various different configurations. Therefore, the detailed description of the embodiments of this application provided below with reference to the accompanying drawings is not intended to limit the scope of protection of the claimed application, but merely represents selected embodiments of this application. All other embodiments obtained by those skilled in the art based on the embodiments of this application without inventive effort are within the scope of protection of this application. The present invention will be further described below with reference to the accompanying drawings.

[0017] like Figure 1 As shown, one embodiment of the present invention relates to an intelligent futures-spot fusion sales execution system, applied to the rebar sales business of a large steel trading company. The system is deployed in an enterprise hybrid cloud environment, with the core modules deployed in a private cloud and the data analysis and machine learning modules deployed in a public cloud, employing a microservice architecture.

[0018] The order management module receives customer orders from multiple channels, including online store orders, telephone orders, and email orders. This module supports three order types: immediate orders (delivery on the same day or the next day), deferred orders (delivery scheduled for a future date), and spot orders (customers select a price point, and the company locks in the price within a specific time window).

[0019] The inventory monitoring module connects in real time to the enterprise's warehouse management systems in multiple major cities across the country.

[0020] The futures-spot analysis module connects to the Shanghai Futures Exchange's market data system via a dedicated line to obtain real-time market data for the main rebar futures contract RB2505, currently priced at 3450 yuan / ton. Simultaneously, it connects to a steel spot price platform, collecting rebar spot prices from 20 major cities nationwide; the current average spot price in Shanghai is 3520 yuan / ton. This module calculates the futures-spot price spread as 3520 - 3450 = 70 yuan / ton, with a spread rate of 70 / 3520 = 1.99%, close to the 2% arbitrage threshold.

[0021] The AI ​​decision engine operates its algorithm based on the current market conditions. Input parameters include: order demand of 80,000 tons, available inventory of 53,000 tons, futures price of 3,450 yuan / ton, spot price of 3,520 yuan / ton, inventory adequacy of 28.75%, and expected demand of 100,000 tons in the next 30 days. The decision engine calculates: futures sales revenue Efutures = 3,450 - 50 (delivery cost) = 3,400 yuan / ton, and spot sales revenue Espot = 3,520 - 80 (logistics + warehousing) = 3,440 yuan / ton. Although the spot revenue is slightly higher, considering that the futures-spot price difference is close to the threshold and inventory is relatively tight, the system recommends a hybrid strategy: 60% through futures sales (establishing short hedging) + 40% through spot sales, while triggering a procurement and restocking plan.

[0022] The risk assessment module comprehensively examined the sales plan. Customer credit check: This month's main customers are large AA-rated construction companies with good credit and no additional guarantees required. Price volatility risk: The price volatility of rebar in the past 30 days (σ=2.3%) is within the normal range and has not triggered a high volatility warning. Inventory risk: After selling 80,000 tons, inventory will drop to -27,000 tons (requiring replenishment through procurement), triggering the emergency procurement process. Hedging ratio: Futures sales of 48,000 tons account for 60% of total sales, which is within a reasonable range. Funding risk: The futures margin requirement is approximately 34.5 million × 48,000 × 15% = 24.84 million yuan, with current available funds of 80 million yuan, indicating ample funds. All risk checks passed.

[0023] The automated execution module began executing sales orders. For the 32,000 tons of spot sales, the system sent an outbound order to the WMS system, allocating inventory based on the nearest customer location: 12,000 tons from the Shanghai warehouse, 10,000 tons from Tianjin, and 10,000 tons from Guangzhou. Simultaneously, a delivery order was generated, and the logistics company was notified to arrange transportation. For the 48,000 tons of futures sales, the system sent sell orders to the futures company via the CTP interface, in 10 separate orders of 480 lots each (4,800 tons), using a counterparty pricing method to ensure rapid execution. The actual average transaction price was 3,451 yuan / ton. The entire execution process took 8 minutes and was fully automated.

[0024] The customer management module records the transaction information and updates the customer profile. This customer's cumulative purchase volume this year has reached 500,000 tons, classifying them as a VIP customer, and the system automatically grants them a 0.5% price discount. The system also marks the customer's payment history; historically, this customer has always settled payments within 30 days of delivery, and their credit score remains at 95 out of 100.

[0025] The data center stores complete data for this transaction, including: transaction time, market snapshot (futures 3450, spot 3520, spread 70, volatility 2.3%), decision parameters (hybrid strategy 60:40), execution results (average spot price of 32,000 tons of 3440 yuan, average futures price of 48,000 tons of 3451 yuan), and estimated profit (3440×3.2 + 3451×4.8 - cost) approximately 275.52 million yuan. This data will be used for subsequent model training and strategy optimization.

[0026] The web-based management backend displays a real-time sales dashboard, showing key metrics such as daily order processing status, inventory trends, and revenue completion rate. The sales director received a system notification for this large transaction via the mobile app and confirmed that the transaction plan was correct.

[0027] Furthermore, such as Figure 6 As shown, the system has a built-in machine learning module, which has the ability to learn on its own and continuously optimize. (1) Feature engineering: Extract feature variables from historical data, including price trend features (technical indicators such as moving average, MACD, RSI, etc.), fundamental features (inventory data, supply and demand relationship, macroeconomic indicators), seasonal features (month, holiday effect), customer features (credit score, purchase frequency, order amount), etc., totaling 80+ features; (2) Model Training: The XGBoost gradient boosting tree algorithm was used to train the sales revenue prediction model and the risk prediction model. The model input is the aforementioned feature vector, and the output is the expected sales revenue and risk probability. The training data comes from historical transaction records of the past 2 years, approximately 10,000+ samples; (3) Model validation: Divide the dataset into training and test sets in an 8:2 ratio, and use cross-validation to evaluate the model performance and ensure the model's generalization ability; (4) Online learning: After each sale is completed, the system compares the actual result with the predicted result, calculates the prediction error, and updates the model parameters through online learning algorithms to achieve continuous optimization of the model; (5) A / B testing: The system regularly conducts strategy comparison tests, randomly assigns orders to different decision-making strategies, evaluates the actual effect of each strategy through statistical analysis, and selects the optimal strategy as the main strategy.

[0028] like Figure 2As shown, another embodiment of the present invention relates to an intelligent futures-spot fusion sales execution method, specifically including the following: Step 1: The system receives customer orders or automatically generates sales plans. Order information includes product specifications, quantity, expected price, delivery date, etc. Step 2: The inventory monitoring module queries the current inventory status to determine if there is enough inventory to meet the order demand. If the inventory is sufficient, proceed to Step 3; if the inventory is insufficient, trigger the procurement process or reject the order. Step 3: The futures and spot analysis module acquires real-time market data, including the futures main contract price P futures, the spot market price P spot, the basis B, and historical price spread series, etc. Step 4: The futures-spot analysis module calculates the futures-spot price spread ratio. The formula is: Price Spread Ratio = (P Futures - P Spot) / P Spot × 100%; Step 5: Determine if the price spread exceeds the threshold δ (usually set at 2%). If it does, there is an arbitrage opportunity, which is marked as a priority futures sale; if it does not exceed the threshold, a decision is made based on a combination of factors. Furthermore, the threshold δ in step 5 is determined based on historical data statistics. Specifically, the method is as follows: calculate the futures-spot price spread rate sequence of the past 180 days, calculate its standard deviation a_history, and set the threshold δ = 1.5 × a_history. This can identify statistically significant price spread opportunities.

[0029] Step 6: As Figure 3 As shown, the AI ​​decision engine initiates multi-objective optimization calculations, and the specific decision logic is as follows: 6.1 Obtain the order quantity Q, current inventory I, safety stock S, and expected demand D; 6.2 Calculate inventory adequacy R = (I - S) / D; 6.3 If R > 1.5, inventory is plentiful, and the tendency is to sell quickly to reduce inventory; if R < 0.8, inventory is tight, and the tendency is to delay sales or raise prices. 6.4 Calculate the futures sales revenue: E_futures = Q × P_futures - Q × (transaction costs + delivery costs); 6.5 Calculate the spot sales revenue Espot = Q × Pspot - Q × (logistics cost + warehousing cost); 6.6 Calculate the comprehensive return index K = α × (E futures - E spot) + β × R - γ × risk coefficient, where α, β, and γ are weighting parameters; 6.7 If K > threshold θ (overall return index is significantly high), select the "futures-first sales strategy" (mainly establishing short positions in the futures market, while replenishing inventory from the spot market to meet physical delivery needs; futures sales typically account for 60%–80%); if K < −θ (overall return index is significantly low), select the "pure spot sales strategy" (all sales are completed through spot channels, without opening futures positions); otherwise (−θ ≤ K ≤ θ), select the balanced mixed strategy (futures sales and spot sales each account for approximately 50%, with both channels executed in tandem). Furthermore, when it is necessary to simultaneously balance the three objectives of maximizing revenue, maximizing inventory turnover, and minimizing risk under multiple constraints such as inventory constraints, capital constraints, and risk constraints, the genetic algorithm iteratively evolves through a large number of candidate sales plans to output the optimal futures / spot sales ratio and quantity, thereby accurately calculating the comprehensive revenue index K in step 6.6, such as... Figure 5 As shown, it specifically includes the following: I. Define the objective function: Objective 1: Maximize sales revenue F1 = Σ(selling price per unit × sales quantity) - Σ(cost); Objective 2: Maximize inventory turnover rate F2 = Annual sales / Average inventory value; Objective 3: Minimize risk exposure F3 = Σ(price risk + credit risk + operational risk).

[0030] II. Define constraints: Constraint 1: Inventory constraint - Sales quantity ≤ Available inventory; Constraint 2: Capital Constraint - Futures margin requirement ≤ Available funds × Safety factor; Constraint 3: Risk Constraint - Sales revenue from a single customer ≤ Total sales revenue × Concentration limit; Constraint 4: Capacity constraint - Daily shipment volume ≤ Logistics processing capacity.

[0031] III. Solve using an improved genetic algorithm: Initialize the population: Randomly generate N sales plans as the initial population, each plan containing decision variables such as sales channel, quantity, price, and time; Fitness assessment: For each solution, calculate three objective function values ​​and evaluate fitness using either the weighted summation method or the Pareto ranking method; Selection process: A tournament selection method is used to select the scheme with high fitness to advance to the next generation; Cross operation: Perform single-point or multi-point cross operations on the selected solutions to generate new sales solutions; Mutation operation: Randomly perturb the decision variables with a certain probability to increase population diversity; Elite retention: Retaining the few solutions with the highest fitness in each generation to ensure algorithm convergence; IV. Iteration Termination Condition: Iteration stops when the number of iterations reaches a preset value (usually 100-200 generations) or the improvement of the objective function is less than the threshold ε (e.g., 0.01%). V. Output the optimal solution: Select the solution with the highest fitness from the final population as the recommended sales strategy.

[0032] Step 7: As Figure 4 As shown, the risk assessment module performs a risk check on the generated sales plan. The check items include: 7.1 Customer Credit Check: Check the customer's credit rating. If the customer is a high-risk customer, require prepayment or a guarantee. 7.2 Price volatility risk: Calculate the historical price volatility σ. If σ > the set threshold, reduce the proportion of futures sales or increase hedging efforts. Price volatility σ is calculated using the GARCH model, with the formula: σt² = ω + α × εt⁻¹² + β ×σt⁻¹², where εt is the price return residual, and ω, α, and β are model parameters obtained through maximum likelihood estimation from historical data.

[0033] 7.3 Inventory Risk: Check whether the post-sale inventory is lower than the safety stock; if so, trigger an emergency procurement plan. 7.4 Hedging Ratio Check: Calculate the proportion of hedging positions to total sales volume and ensure it is within a reasonable range (usually 50%-100%). 7.5 Funding Risk: Check whether the futures margin requirement exceeds the safe ratio of available funds; Step 8: If any risk test fails, adjust the sales plan or issue a risk warning, which requires manual review; if all tests pass, proceed to Step 9. Step 9: The automatic execution module executes the sales instruction. Specific operations include: 9.1 If spot sales are selected, an outbound instruction will be sent to the WMS system to generate a delivery note and notify the logistics department to arrange delivery. 9.2 If futures sales are selected, a sell order is sent to the futures trading system to establish a short hedging position. 9.3 If a mixed strategy is adopted, spot and futures operations shall be executed in proportion to each other. 9.4 Simultaneously update the inventory system and lock the corresponding inventory to prevent duplicate sales; Step 10: The system monitors the order execution status in real time, tracking the progress of spot delivery, futures transaction status, and customer payment status; Step 11: Once the spot goods are delivered or the futures position is closed, the system automatically calculates the profit, compares the actual profit with the expected profit, and analyzes the reasons for the deviation. Step 12: Record the complete sales data in the data center, including market status, decision parameters, execution results, and revenue, for subsequent model optimization; Step 13: Determine whether to continue monitoring the market and processing new orders. If yes, return to Step 1; otherwise, end the process.

[0034] The above description is merely a preferred embodiment of the present invention. It should be understood that the present invention is not limited to the forms disclosed herein and should not be construed as excluding other embodiments. It can be used in various other combinations, modifications, and improvements, and can be altered within the scope of the concept described herein through the above teachings or related technologies or knowledge. Modifications and variations made by those skilled in the art that do not depart from the spirit and scope of the present invention should be within the protection scope of the appended claims.

Claims

1. An intelligent futures-spot integrated sales execution system, characterized in that: The system includes an order management module, an inventory monitoring module, a futures and spot analysis module, an AI decision engine, a risk assessment module, and an automatic execution module. The order management module is configured to receive and process customer information. The inventory monitoring module is configured to track the enterprise's inventory status in real time. The futures-spot analysis module is configured to collect futures market data and spot market quotes in real time, calculate the futures-spot price difference, basis, and arbitrage opportunities, and analyze market trends and price movements. The AI ​​decision engine is configured to generate the optimal sales plan based on information such as order demand, inventory status, futures and spot price relationships, and market expectations, using optimization algorithms. The risk assessment module is configured to perform a multi-dimensional risk assessment of the sales plan. The automatic execution module is configured to automatically execute spot delivery instructions, futures trading instructions, and inventory transfer instructions based on the sales plan generated by the decision engine, and monitor the execution status in real time.

2. The intelligent futures-spot fusion sales execution system according to claim 1, characterized in that: The system also includes a customer management module and a data center; The customer management module is configured to maintain a customer information database, including customer credit registration, historical transaction records, payment habits, and preferred products, providing customer profiles for sales decisions. The data center is configured to use a combination of time-series and relational databases to store historical market data, order data, inventory data, and transaction records to support big data analytics and machine learning model training.

3. The intelligent futures-spot fusion sales execution system according to claim 1, characterized in that: The method of using optimization algorithms to generate the optimal sales plan includes: Obtain the order quantity Q, current inventory I, total inventory S, and expected demand D; Calculate the inventory adequacy R = (IS) / D. If R > 1.5, it indicates that the inventory is sufficient and you tend to sell quickly to reduce inventory. If R is less than 0.8, it indicates that the inventory is tight and you tend to delay sales or raise prices. Calculate the futures sales revenue E_futures = Q × P_futures - Q × (transaction costs + delivery costs); Calculate the spot sales revenue Espot = Q × Pspot - Q × (logistics cost + warehousing cost); The comprehensive return index K is calculated as follows: K = α × (E futures - E spot) + β × R - γ × risk coefficient, where α, β, and γ are weighting parameters. If K > threshold θ, then choose the futures sales + spot replenishment strategy with futures sales as the priority; if K < -θ, then choose the pure spot sales strategy; otherwise, choose the mixed strategy.

4. The intelligent futures-spot fusion sales execution system according to claim 1, characterized in that: The multi-dimensional risk assessment includes: Customer credit check: Check the customer's credit record. If the requirements are not met, a prepayment or guarantee is required. Price volatility risk: Calculate the historical price volatility σ. If σ > a set threshold, reduce the proportion of futures sales or increase hedging efforts. Inventory risk: After the price difference is sold, is the inventory below the safety stock? If so, an emergency procurement plan will be triggered. Hedging ratio check: Calculate the proportion of hedging positions to total sales volume and ensure it is within the set range; Funding risk: Check whether the futures margin required exceeds the safe ratio of available funds.

5. A method based on an intelligent futures-spot fusion sales execution system according to any one of claims 1-4, characterized in that: The method includes: S1. When the system receives customer orders or automatically generates sales plans, it performs spot-futures analysis when the inventory monitoring module finds that there is enough inventory to meet the order demand, and determines whether the price difference rate exceeds the threshold δ. If it does, it is marked as a priority futures sale. S2. The AI ​​decision engine initiates multi-objective optimization calculations to obtain the comprehensive return index K. If K > threshold θ, then the futures sales + spot replenishment strategy with futures sales priority is selected. If K < -θ, then the pure spot sales strategy is selected. Otherwise, the mixed strategy is selected. S3. Conduct risk checks on the generated sales plan. Once all risk checks are passed, execute the sales order and monitor the order execution status in real time, tracking the progress of spot delivery, futures transactions, and customer payment status.

6. The method based on an intelligent futures-spot fusion sales execution system according to claim 5, characterized in that: The method further includes: S4. Once the spot goods are delivered or the futures position is closed, the profit calculation is performed automatically, the actual profit is compared with the expected profit, and the reasons for the deviation are analyzed. S5. Record the complete sales data to the data center and determine whether to continue monitoring the market and processing new orders.

7. The method based on an intelligent futures-spot fusion sales execution system according to claim 5, characterized in that: The AI ​​decision engine initiates multi-objective optimization calculations to obtain the comprehensive return index K, which includes: Obtain the order quantity Q, current inventory I, total inventory S, and expected demand D; Calculate the inventory adequacy R = (IS) / D. If R > 1.5, it indicates that the inventory is sufficient and you tend to sell quickly to reduce inventory. If R is less than 0.8, it indicates that the inventory is tight and you tend to delay sales or raise prices. Calculate the futures sales revenue E_futures = Q × P_futures - Q × (transaction costs + delivery costs); Calculate the spot sales revenue Espot = Q × Pspot - Q × (logistics cost + warehousing cost); The comprehensive return index K is calculated as follows: K = α × (E futures - E spot) + β × R - γ × risk coefficient, where α, β, and γ are weighting parameters.

8. The method based on an intelligent futures-spot fusion sales execution system according to claim 5, characterized in that: The risk assessment includes: Customer credit check: Check the customer's credit record. If the requirements are not met, a prepayment or guarantee is required. Price volatility risk: Calculate the historical price volatility σ. If σ > a set threshold, reduce the proportion of futures sales or increase hedging efforts. Inventory risk: After the price difference is sold, is the inventory below the safety stock? If so, an emergency procurement plan will be triggered. Hedging ratio check: Calculate the proportion of hedging positions to total sales volume and ensure it is within the set range; Funding risk: Check whether the futures margin required exceeds the safe ratio of available funds.

9. The method based on an intelligent futures-spot fusion sales execution system according to claim 5, characterized in that: The execution of the sales instruction includes: If spot sales are selected, an outbound instruction is sent to the WMS system to generate a delivery note and notify the logistics department to arrange delivery. If you choose to sell futures, send a sell order to the futures trading system to establish a short hedging position. If a mixed strategy is adopted, spot and futures operations will be executed in proportion to each other. Update the inventory system and lock the corresponding inventory to prevent duplicate sales.

Images