Intelligent purchase and inventory early warning method and system for catering raw materials

Abstract

The invention relates to the technical field of catering management, in particular to a catering raw material intelligent purchasing and inventory early warning method and system, and the method comprises the steps: obtaining current raw material inventory data, order data and historical sales data; acquiring historical price data of the raw materials; predicting a raw material demand based on the current raw material inventory data, the order data and the historical sales volume data; predicting a raw material price fluctuation trend based on the raw material historical price data; determining an optimal purchasing opportunity and an optimal purchasing quantity according to the predicted raw material demand quantity and the predicted price fluctuation trend; inventory early warning calculation is carried out based on the arrival time, the shelf life, the demand quantity and the storage environment parameters of the raw materials; generating purchase suggestion information and inventory early warning information; and pushing the purchase suggestion information and the inventory early warning information to a management terminal. According to the method, demand prediction and price trend analysis are carried out by adopting an innovative multi-model integration method, so that the prediction accuracy and stability are greatly improved.

Description

Technical Field

[0001] The present invention relates to the technical field of catering management, and more specifically, to a method and system for intelligent procurement and inventory early warning of catering raw materials. Background Art

[0002] In recent years, with the rapid development of the catering industry and the continuous improvement of consumers' requirements for food quality, catering companies are facing increasing challenges in raw material procurement and inventory management. Traditional catering raw material procurement and inventory management methods mainly rely on manual experience and simple data statistics, which is no longer sufficient in the modern competitive catering market.

[0003] Currently, there are some procurement and inventory management systems for the catering industry on the market, but these systems still have many shortcomings. First, most systems can only provide basic inventory monitoring and simple demand forecasting, and cannot effectively respond to the rapid changes in market demand and fluctuations in food prices. Secondly, existing systems often ignore the particularity of raw materials in the catering industry, such as perishability and seasonality, resulting in poor inventory management and serious food waste. Furthermore, the supplier selection and evaluation process is still highly dependent on manual decision-making, lacks scientific quantitative standards, and is easily affected by subjective factors.

[0004] In addition, existing technologies also have obvious deficiencies in information integration and decision support. Most systems cannot effectively integrate data from multiple links such as procurement, inventory, and sales, making it difficult to provide managers with comprehensive and timely decision support. At the same time, the information push mechanism is not intelligent enough to provide personalized information services based on the needs of different roles.

[0005] The closest existing technology usually uses simple time series analysis for demand forecasting and fixed thresholds for inventory warning. This method is difficult to adapt to the complex and changing business environment of the catering industry. In terms of purchasing decisions, existing technologies often only consider price factors and ignore important dimensions such as supplier evaluation and delivery time, resulting in a lack of comprehensiveness in purchasing decisions.

[0006] In view of the above problems, the catering industry is in urgent need of a more intelligent, comprehensive and efficient raw material procurement and inventory warning solution. The present invention is proposed to meet this demand and aims to solve many problems existing in the prior art, such as inaccurate demand forecasting, improper inventory management leading to food waste, lack of scientific basis for procurement decisions, and insufficient timeliness and personalization of information push. Summary of the invention

[0007] The present invention proposes an innovative intelligent procurement and inventory early warning method and system for catering raw materials by comprehensively applying a variety of advanced data analysis and artificial intelligence technologies. The method and system can significantly improve procurement efficiency, optimize inventory management, reduce operating costs, and ensure the freshness and timely supply of ingredients.

[0008] The present invention provides a method for intelligent procurement and inventory early warning of catering raw materials, comprising:

[0009] The acquisition steps include:

[0010] Obtain current raw material inventory data, order data, and historical sales data;

[0011] Get historical price data of raw materials;

[0012] Processing steps include:

[0013] Based on the current raw material inventory data, order data and historical sales data, forecast the demand for raw materials;

[0014] Based on the historical price data of raw materials, predict the price fluctuation trend of raw materials;

[0015] Determine the optimal purchase time and purchase quantity based on the predicted raw material demand and price fluctuation trend;

[0016] Calculate inventory warning based on raw material arrival time, shelf life, demand and storage environment parameters;

[0017] Output steps include:

[0018] Generate purchase suggestion information and inventory warning information;

[0019] The purchase suggestion information and inventory warning information are pushed to the management terminal.

[0020] Preferably, the step of predicting the demand for raw materials specifically includes:

[0021] Building a demand forecasting model based on the historical sales data;

[0022] Inputting the current raw material inventory data and order data into the demand forecasting model;

[0023] The short-term demand forecast results and the long-term demand forecast results are obtained, where the short-term is 1-3 days and the long-term is 3-7 days.

[0024] Preferably, the step of predicting the price fluctuation trend of raw materials specifically includes:

[0025] Based on the historical price data of raw materials, a price prediction model is constructed;

[0026] At least one of a linear fitting model, an exponential fitting model, a moving average model, a time series autoregression model or a Markov chain model is used as the price prediction model.

[0027] Preferably, the step of determining the optimal purchase time and purchase quantity specifically includes:

[0028] Based on the predicted price fluctuation trend, determine the lowest price point as the purchase time;

[0029] Calculate the optimal purchase quantity based on the predicted raw material demand and raw material shelf life;

[0030] Determine the actual purchase quantity based on the minimum batch size provided by the supplier.

[0031] Preferably, the step of performing inventory warning calculation specifically includes:

[0032] Calculate the preset period storage volume of raw materials, the calculation formula is:

[0033]

[0034] Among them, e is the preset period storage quantity, S is the preset raw material unit net content, b is the storage environment coefficient, y is the storage period, a is the minimum batch provided by the supplier, r is the weekly sales volume, T is the raw material arrival time, and t is the time since the last order;

[0035] The preset period storage quantity is compared with the actual storage quantity to obtain an inventory warning result.

[0036] As a preference, a supplier selection step is also included:

[0037] Obtain supplier evaluation data, supplier distance data and historical supply data;

[0038] Calculate supplier ratings based on the supplier evaluation data, supplier distance data, and historical supply data;

[0039] Based on the supplier ratings, the best supplier is selected.

[0040] Preferably, the step of calculating the supplier rating adopts the following formula:

[0041]

[0042] Among them, P is the supplier rating, P1 and P2 are weight coefficients, x is the average supplier evaluation score, D is the supplier distance, and D min is the minimum distance among all suppliers.

[0043] As a preferred embodiment, the step of calculating the degree of order dispersion is also included:

[0044] Based on the historical sales data, calculating sales variance;

[0045] The inverse of the sales volume variance is used as an indicator of order dispersion;

[0046] The order dispersion index is used as one of the reference bases for supplier selection.

[0047] Preferably, the step of pushing the purchase suggestion information and inventory warning information to the management terminal specifically includes:

[0048] Generate push messages containing the best purchasing time, purchasing quantity, inventory warning results and the best supplier information;

[0049] The push message is sent to the mobile terminal of the manager via a mobile application or enterprise WeChat.

[0050] Intelligent procurement and inventory early warning system for catering raw materials, including:

[0051] The acquisition module is used to obtain current raw material inventory data, order data, historical sales data and historical raw material price data;

[0052] A demand forecasting module, used to forecast the demand for raw materials based on the current raw material inventory data, order data and historical sales data;

[0053] A price prediction module, used to predict the price fluctuation trend of raw materials based on the historical price data of the raw materials;

[0054] A purchasing decision module, used to determine the optimal purchasing time and purchase quantity according to the predicted raw material demand and price fluctuation trend;

[0055] Inventory warning module, used to calculate inventory warning based on raw material arrival time, shelf life, demand and storage environment parameters;

[0056] Supplier selection module, used to select the best supplier based on supplier evaluation data, supplier distance data and historical supply data;

[0057] The information push module is used to generate purchase suggestion information and inventory warning information, and push the purchase suggestion information and inventory warning information to the management terminal.

[0058] The beneficial effects of the present invention are mainly reflected in the following aspects:

[0059] First, the invention adopts an innovative multi-model integration method for demand forecasting and price trend analysis, which greatly improves the accuracy and stability of the forecast. This method can effectively capture the seasonal fluctuations and sudden changes in demand in the catering industry, and provide reliable data support for purchasing decisions.

[0060] Secondly, the inventory warning algorithm proposed by the present invention fully considers the particularity of catering raw materials, such as shelf life, storage conditions and other factors, and realizes more accurate inventory management. This not only reduces food waste, but also ensures the freshness of food, thereby improving catering quality.

[0061] Furthermore, the present invention introduces an innovative scoring model in the supplier selection process, which comprehensively considers multiple dimensions such as quality, price, and delivery time, making the supplier selection more scientific and reasonable. This method helps to establish a stable and efficient supply chain network and reduce procurement risks.

[0062] In addition, the information push mechanism of the present invention has highly intelligent and personalized features. The system can provide customized information services according to the responsibilities and permissions of different managers, ensuring that key information can be conveyed to relevant decision makers in a timely and accurate manner.

[0063] Finally, the modular design of the present invention not only improves the maintainability and scalability of the system, but also achieves seamless connection of each link. This design makes the entire procurement and inventory management process smoother and more efficient, significantly improving the overall operational efficiency of the catering enterprise.

[0064] In general, the present invention realizes scientific purchasing decision-making, refined inventory management and intelligent information service through innovation and optimization of various links. These innovations not only solve many problems existing in the prior art, but also provide a comprehensive, efficient and reliable raw material management solution for catering enterprises, which strongly supports the sustainable development and competitiveness improvement of catering enterprises. BRIEF DESCRIPTION OF THE DRAWINGS

[0065] Figure 1 The figure is a flow chart of the method of the present invention.

[0066] Figure 2 It is a logic block diagram of the demand forecasting module of the present invention.

[0067] Figure 3 It is a logic block diagram of the price prediction module of the present invention.

[0068] Figure 4 It is a logic block diagram of the purchasing decision module of the present invention.

[0069] Figure 5 It is a logic block diagram of the demand forecasting module of the present invention.

[0070] Figure 6 It is a logic block diagram of the price prediction module of the present invention.

[0071] Figure 7 It is a logic block diagram of the purchasing decision module of the present invention. DETAILED DESCRIPTION

[0072] Please refer to Figure 1-7 The present invention provides a method and system for intelligent procurement and inventory warning of catering raw materials. The method aims to solve the pain points of raw material procurement and inventory management in the catering industry, improve procurement efficiency, reduce inventory costs, and ensure the freshness of ingredients.

[0073] First, the method includes an acquisition step. In this step, the system acquires current raw material inventory data, order data, and historical sales data. These data are the basis for making intelligent purchasing decisions. Preferably, the system also acquires historical price data of raw materials for price trend analysis. For example, for a medium-sized chain restaurant, the system may acquire daily sales data for the past three months, current inventory of various ingredients, pre-order information for the next week, and price fluctuation data of major ingredients such as meat and vegetables in the past six months.

[0074] Next, the method enters the processing step. This step is the core of the present invention and includes multiple sub-steps. First, based on the current raw material inventory data, order data and historical sales data obtained, the system predicts the demand for raw materials. In one embodiment of the present invention, the system uses a time series analysis method to predict demand. For example, an ARIMA (autoregressive integrated moving average) model can be used, and the mathematical expression of the model is as follows:

[0075] φ(B)(1-B) d X t =θ(B)ε t ,

[0076] Where B is the lag operator, d is the number of differences, φ(B) and θ(B) are the autoregressive and moving average polynomials, respectively, and X t is the time series, ε t is a white noise process.

[0077] Through this model, the system can predict the demand for various raw materials in the next week. Preferably, the system will make short-term (1-3 days) and long-term (3-7 days) demand forecasts separately to make better purchasing decisions.

[0078] Secondly, the system predicts the price fluctuation trend of raw materials based on the historical price data of raw materials. In another embodiment of the present invention, the system uses a support vector regression (SVR) model for price prediction. The objective function of the SVR model can be expressed as:

[0079]

[0080] st

[0081]

[0082] Among them, w is the weight vector, b is the bias term, C is the penalty parameter, ε is the insensitive loss function value, ξ i and is the slack variable.

[0083] Through this model, the system can predict the price trend of various raw materials in the future. For example, for seasonal fruits and vegetables, the system may predict that the price will drop significantly in a certain period of time, thus providing an important reference for purchasing decisions.

[0084] Then, based on the predicted demand for raw materials and price fluctuation trends, the system determines the optimal purchase time and purchase quantity. This step involves a multi-objective optimization problem, which requires balancing procurement costs, inventory costs, and out-of-stock risks. In one embodiment of the present invention, the system uses a genetic algorithm to solve this optimization problem. The basic process of the genetic algorithm is as follows: 1. Initialize the population; 2. Evaluate fitness; 3. Select; 4. Crossover; 5. Mutation; 6. Repeat steps 2-5 until the termination condition is met; Through the genetic algorithm, the system can find a procurement plan that minimizes the total cost while meeting the demand.

[0085] Finally, the system calculates inventory warnings based on the arrival time, shelf life, demand, and storage environment parameters of the raw materials. In this step, the system considers multiple factors, such as the perishability of the raw materials and the impact of storage conditions on the shelf life. For example, for meat that needs to be refrigerated, the system adjusts the warning threshold based on the temperature parameters of the refrigeration equipment. Preferably, the system uses fuzzy logic control to achieve inventory warnings. The basic steps of fuzzy logic control include:

[0086] 1. Fuzzification; 2. Fuzzy reasoning; 3. Defuzzification; Through this method, the system can handle various uncertainties more flexibly and improve the accuracy of inventory warning.

[0087] In the output step, the system generates purchase suggestion information and inventory warning information, and pushes this information to the management terminal. This information may include the type, quantity, and time of the recommended purchase of raw materials, as well as which raw materials are about to run out of stock or approach their expiration dates. Preferably, the system will push this information through mobile applications or instant messaging tools such as WeChat for Enterprise to ensure that managers can obtain and make decisions in a timely manner.

[0088] Through the above steps, the method of the present invention can significantly improve the raw material procurement efficiency of catering enterprises, reduce inventory costs, reduce food waste, and ensure the freshness and timely supply of food. In particular, for chain catering enterprises operating multiple stores, this method can achieve cross-store inventory optimization and further improve overall operational efficiency.

[0089] Next, several key technical points of the present invention and their implementation methods will be introduced in detail. In a preferred embodiment of the present invention, the step of predicting the demand for raw materials is further refined. First, the system builds a demand forecasting model based on historical sales data. The choice of this model is crucial because it directly affects the accuracy of the forecast. This method preferably uses the long short-term memory network (LSTM) in deep learning as a forecasting model, because LSTM is particularly good at processing time series data and can capture long-term dependencies in sales data.

[0090] After the model training is completed, the system inputs the current raw material inventory data and order data into the LSTM model. The purpose of this step is to enable the model to make predictions based on the latest business conditions and improve the timeliness and accuracy of the predictions. For example, if the system detects that there are large-scale event reservations in the near future, it will adjust the demand forecast accordingly.

[0091] Finally, the system obtains short-term demand forecast results and long-term demand forecast results. In the present invention, short-term is defined as 1-3 days, and long-term is defined as 3-7 days. This hierarchical forecasting method can meet the needs of daily restaurant operations and medium- and long-term planning at the same time. For example, short-term forecasts can be used to guide daily purchasing decisions, while long-term forecasts can be used to formulate weekly purchasing plans or negotiate prices with suppliers.

[0092] When predicting the price fluctuation trend of raw materials, the present invention adopts a multi-model integration method to improve the stability and accuracy of the prediction. Specifically, the system will use a linear fitting model, an exponential fitting model, a moving average model, a time series autoregressive model and a Markov chain model at the same time. The prediction results of each model will be given different weights according to its historical performance, and finally a comprehensive prediction result will be obtained.

[0093] Taking the time series autoregression model (AR model) as an example, its mathematical expression is as follows:

[0094]

[0095] Among them, X t is the price at time t, c is a constant term, φ i is the autoregressive coefficient, p is the autoregressive order, ε t is white noise.

[0096] Through this multi-model integration method, the present invention can effectively capture the price fluctuation characteristics of different types of raw materials, thereby providing more reliable price forecast information for procurement decisions.

[0097] In the step of determining the optimal purchase time and purchase quantity, the present invention adopts an innovative multi-objective optimization algorithm. First, based on the predicted price fluctuation trend, the system identifies the possible lowest price point as a potential purchase opportunity. The peak and valley detection algorithm is used here, and its core idea is to find the local minimum point of the price curve.

[0098] Then, the system calculates the optimal purchase quantity based on the predicted raw material demand and the shelf life of the raw materials. This calculation process takes into account multiple factors, including storage cost, out-of-stock cost, procurement cost, etc. Preferably, the system uses a dynamic programming algorithm to solve this optimization problem. Let f(t,q) be the minimum total cost starting from day t when the current inventory is q, then its state transition equation can be expressed as:

[0099]

[0100] Among them, a is the purchase quantity, c(q,a) is the cost function of the current decision, and d t is the demand on day t.

[0101] Finally, the system will also determine the actual purchase quantity based on the minimum batch size provided by the supplier. This is because in actual operations, suppliers usually set a minimum order quantity. The system will adjust the calculated optimal purchase quantity upward to the closest batch unit to meet the supplier's requirements.

[0102] Through this method, the present invention can find the best balance between meeting demand, controlling costs and complying with supplier rules, thereby realizing truly intelligent purchasing decisions. In a preferred embodiment of the present invention, the inventory warning calculation step adopts an innovative method that fully considers the particularity of raw material management in the catering industry. First, the system calculates the preset period storage volume of raw materials. This calculation process integrates multiple key factors, making the warning more accurate and reliable.

[0103] Specifically, the calculation formula for the preset cycle storage amount is as follows:

[0104]

[0105] In this formula, e represents the preset cycle storage quantity, S represents the preset unit net content of raw materials, b is the storage environment coefficient, q represents the storage life, a represents the minimum batch provided by the supplier, r is the weekly sales volume, T represents the arrival time of raw materials, and t is the time since the last order.

[0106] For perishable fresh ingredients, the P value will be larger, resulting in a smaller preset storage capacity; while for dry food ingredients, the b value will be smaller, allowing for a larger storage capacity. In actual applications, the b value may be dynamically adjusted according to seasonal changes. For example, in summer, the b value of fresh vegetables will be higher than in winter.

[0107] At the same time, the formula also takes into account the supplier's minimum batch a, which reflects the in-depth understanding of the actual procurement operation of the present invention. Many suppliers will set a minimum order quantity, and ignoring this may cause the inventory warning system to be out of touch with the actual procurement operation. By incorporating a into the calculation, the present invention ensures the practicality of the warning system.

[0108] After calculating the preset period storage volume, the system will compare it with the actual storage volume to obtain the inventory warning result. This comparison process is not a simple numerical comparison, but a fuzzy logic control method. For example, the system may set multiple warning levels, such as "sufficient", "attention", "warning" and "urgent". Each level corresponds to a fuzzy set, and the ratio of the actual storage volume to the preset storage volume is used as input, and the final warning level is obtained through fuzzy reasoning.

[0109] The advantage of this approach is that it can handle uncertainty and ambiguity in inventory management. For example, for some ingredients with a short shelf life, even if the actual storage volume is slightly higher than the preset storage volume, the system may issue a "caution" level warning to remind managers to pay attention to the use of these ingredients.

[0110] In another embodiment of the present invention, the system also includes a supplier selection step, which is a key link in intelligent purchasing decisions. First, the system obtains supplier evaluation data, supplier distance data, and historical supply data. These data sources may include subjective evaluations of restaurant employees, delivery times recorded by the logistics system, and fulfillment of past orders.

[0111] Next, the system calculates the supplier score based on this data. This scoring process uses a comprehensive evaluation model, and its mathematical expression is as follows:

[0112]

[0113] In this formula, P represents the final score of the supplier, P1 and P2 are weight coefficients, reflecting the importance of quality evaluation and distance factors respectively. x represents the average evaluation score of the supplier (assuming the full score is 5 points), D is the actual distance of the supplier, and D min is the minimum distance among all candidate suppliers.

[0114] The design concept of this scoring model is to balance quality and convenience. By adjusting the values ​​of P1 and P2, restaurants can emphasize quality or convenience according to their own business strategies. For example, for a restaurant that focuses on high-end ingredients, a larger P1 value may be set to ensure that the best quality raw materials are obtained; while for a fast food chain, the weight of P2 may be increased to reduce logistics costs and improve supply efficiency.

[0115] It is worth noting that this scoring model can be further refined. For example, more evaluation dimensions can be introduced, such as the supplier's price level, product diversity, after-sales service quality, etc. Each dimension can be given a different weight to build a more comprehensive scoring system.

[0116] Finally, the system will select the best supplier based on the calculated supplier scores. This selection process is not just about simply selecting the supplier with the highest score, but also takes other factors into consideration, such as whether the supplier's production capacity can meet current procurement needs and whether there are strategic considerations for long-term cooperation.

[0117] Through this scientific supplier selection method, the present invention can help catering companies establish a stable, efficient, and high-quality supply chain network, thereby improving overall operational efficiency and food quality. In an innovative embodiment of the present invention, an order dispersion calculation step is introduced, which is an important supplement to the traditional supplier selection method. First, the system calculates the sales variance based on historical sales data. The purpose of this step is to quantify the stability of restaurant orders, because the stability of orders directly affects the management efficiency of the supply chain.

[0118] Specifically, the sales volume variance is calculated as follows:

[0119]

[0120] Among them, σ 2 represents the sales variance, n is the number of samples, x i is the sales volume on day i, is the average sales volume. Next, the system uses the inverse of the sales volume variance as an indicator of order dispersion. The mathematical expression of this indicator is:

[0121]

[0122] Among them, D represents the order dispersion index. The cleverness of this method is that it can effectively reflect the stability of orders. When sales fluctuate greatly, the variance will increase, resulting in a smaller D value, indicating a high degree of order dispersion; conversely, when sales are relatively stable, the variance is small and the D value is large, indicating a high degree of order concentration.

[0123] The present invention uses this order dispersion index as one of the important references for supplier selection. This is because different suppliers may have different advantages in handling stable orders and fluctuating orders. For example, in the case of high order dispersion, the system may tend to select suppliers that can provide flexible supply services; while in the case of high order concentration, the system may give priority to suppliers that can provide bulk discounts.

[0124] In practical applications, the order dispersion index can be used in conjunction with the supplier rating model mentioned above. For example, it can be used as an additional rating dimension or used to adjust the weight of supplier ratings. This approach can make supplier selection more accurate and better suited to the actual operating conditions of the restaurant.

[0125] In another preferred embodiment of the present invention, the information push step is further optimized. First, the system generates a push message containing the best purchasing time, purchasing quantity, inventory warning results and the best supplier information. The generation process of this message is not a simple data listing, but has undergone intelligent processing.

[0126] For example, when recommending the timing of purchases, the system will combine price forecast results and inventory status to give specific suggestions, such as "It is recommended to purchase rice within 3 days, which is expected to save 5% of the cost." For the purchase volume, the system will give a reasonable range instead of a fixed value, so that managers can make fine adjustments based on actual conditions. The inventory warning results will be presented in a striking way, with a brief analysis of the reasons, such as "Potato inventory is low (current inventory is only enough for 3 days), and prices are on the rise recently."

[0127] It is worth mentioning that these push messages are also personalized according to the responsibilities and permissions of different managers. For example, a purchasing manager may receive detailed price analysis and supplier comparison information, while a chef may pay more attention to inventory warnings and food freshness information.

[0128] Next, the system will send these push messages to the manager's mobile terminal through a mobile application or WeChat for Business. This instant push mechanism ensures the timeliness of information, allowing managers to make timely decisions. Preferably, the system will also set different push frequencies and urgency levels. For example, daily inventory status may be pushed once a day, while urgent out-of-stock warnings may be pushed immediately with a sound reminder.

[0129] In addition, the system of the present invention also supports interactive information query. Managers can query more detailed data and analysis results at any time through mobile applications, such as historical price trend charts, supplier evaluation details, etc. This flexible way of obtaining information further improves the practicality and user-friendliness of the system.

[0130] Finally, the present invention also provides a complete intelligent procurement and inventory early warning system for catering raw materials. The system includes multiple functional modules, each module focuses on a specific task, and together constitutes a comprehensive solution.

[0131] Specifically, the system includes:

[0132] Acquisition module 1 is responsible for acquiring current raw material inventory data, order data, historical sales data and historical raw material price data.

[0133] Demand forecasting module 2 predicts the demand for raw materials based on the acquired data.

[0134] Price forecasting module 3 analyzes historical price data and predicts the price fluctuation trend of raw materials.

[0135] The purchasing decision module 4 determines the optimal purchasing time and purchase quantity according to the demand forecast and price forecast results.

[0136] The inventory warning module 5 considers various factors and performs inventory warning calculations.

[0137] Supplier selection module 6, evaluate and select the best suppliers.

[0138] Information push module 7 generates and pushes purchase suggestions and inventory warning information.

[0139] These modules are closely connected through data flow and control flow to form an organic whole. For example, the output of the demand forecasting module will directly affect the operation of the procurement decision module, and the results of the inventory warning module will be promptly communicated to the management personnel through the information push module.

[0140] Through this modular design, the system not only achieves functional decoupling and code maintainability, but also leaves room for future expansion and optimization. For example, new prediction algorithms can be easily integrated or new decision dimensions can be added without affecting the stability of the entire system.

[0141] In general, the intelligent procurement and inventory early warning method and system for catering raw materials provided by the present invention provides a comprehensive, accurate and efficient raw material management solution for catering enterprises by comprehensively using a variety of advanced data analysis and artificial intelligence technologies. It can not only significantly improve procurement efficiency and reduce inventory costs, but also ensure the freshness of ingredients and the timeliness of supply, thus making an important contribution to improving catering quality and operating efficiency.

[0142] It should be noted that the above description is only a preferred embodiment of the present invention and is not intended to limit the present invention. Any modification, replacement, and improvement made within the principles of the present invention should be included in the protection scope of the present invention.

Claims

1. Intelligent procurement and inventory early warning method for catering raw materials, characterized in that: include: The acquisition steps include: Obtain current raw material inventory data, order data, and historical sales data; Get historical price data of raw materials; Processing steps include: Based on the current raw material inventory data, order data and historical sales data, forecast the demand for raw materials; Based on the historical price data of raw materials, predict the price fluctuation trend of raw materials; Determine the optimal purchase time and purchase quantity based on the predicted raw material demand and price fluctuation trend; Calculate inventory warning based on raw material arrival time, shelf life, demand and storage environment parameters; Output steps include: Generate purchase suggestion information and inventory warning information; The purchase suggestion information and inventory warning information are pushed to the management terminal.

2. The method according to claim 1, characterized in that The step of predicting the demand for raw materials specifically includes: Building a demand forecasting model based on the historical sales data; Inputting the current raw material inventory data and order data into the demand forecasting model; The short-term demand forecast results and the long-term demand forecast results are obtained, where the short-term is 1-3 days and the long-term is 3-7 days.

3. The method according to claim 1, characterized in that: The step of predicting the price fluctuation trend of raw materials specifically includes: Based on the historical price data of raw materials, a price prediction model is constructed; At least one of a linear fitting model, an exponential fitting model, a moving average model, a time series autoregression model or a Markov chain model is used as the price prediction model.

4. The method according to claim 1, characterized in that: The steps of determining the optimal purchase time and purchase quantity specifically include: Based on the predicted price fluctuation trend, determine the lowest price point as the purchase time; Calculate the optimal purchase quantity based on the predicted raw material demand and raw material shelf life; Determine the actual purchase quantity based on the minimum batch size provided by the supplier.

5. The method according to claim 1, characterized in that The steps of performing inventory warning calculation specifically include: Calculate the preset period storage volume of raw materials, the calculation formula is: Among them, e is the preset period storage quantity, S is the preset raw material unit net content, b is the storage environment coefficient, y is the storage period, a is the minimum batch provided by the supplier, r is the weekly sales volume, T is the raw material arrival time, and t is the time since the last order; The preset period storage quantity is compared with the actual storage quantity to obtain an inventory warning result.

6. The method according to claim 1, characterized in that It also includes the supplier selection step: Obtain supplier evaluation data, supplier distance data and historical supply data; Calculate supplier ratings based on the supplier evaluation data, supplier distance data, and historical supply data; Based on the supplier ratings, the best supplier is selected.

7. The method according to claim 6, characterized in that The steps for calculating supplier ratings use the following formula: Among them, P is the supplier rating, P1 and P2 are weight coefficients, x is the average supplier evaluation score, D is the supplier distance, and D min is the minimum distance among all suppliers.

8. The method according to claim 1, characterized in that It also includes the order dispersion calculation steps: Based on the historical sales data, calculating sales variance; The inverse of the sales volume variance is used as an indicator of order dispersion; The order dispersion index is used as one of the reference bases for supplier selection.

9. The method according to claim 1, characterized in that: The step of pushing the purchase suggestion information and inventory warning information to the management terminal specifically includes: Generate push messages containing the best purchasing time, purchasing quantity, inventory warning results and the best supplier information; The push message is sent to the mobile terminal of the manager via a mobile application or enterprise WeChat.

10. Intelligent procurement and inventory warning system for catering raw materials, characterized by: include: The acquisition module is used to obtain current raw material inventory data, order data, historical sales data and historical raw material price data; A demand forecasting module, used to forecast the demand for raw materials based on the current raw material inventory data, order data and historical sales data; A price prediction module, used to predict the price fluctuation trend of raw materials based on the historical price data of the raw materials; A purchasing decision module, used to determine the optimal purchasing time and purchase quantity according to the predicted raw material demand and price fluctuation trend; Inventory warning module, used to calculate inventory warning based on raw material arrival time, shelf life, demand and storage environment parameters; Supplier selection module, used to select the best supplier based on supplier evaluation data, supplier distance data and historical supply data; The information push module is used to generate purchase suggestion information and inventory warning information, and push the purchase suggestion information and inventory warning information to the management terminal.

Images