KR-102964696-B1 - Integrated Management System for Sale Optimization and Revenue Enhancement based on Artificial Intelligence

Abstract

The present invention relates to an integrated management system for artificial intelligence-based sales optimization and revenue growth, and is a technology designed to support essential sales activities for improving revenue, particularly for small and medium-sized travel agencies and small and medium-sized distributors.

Inventors

• 박이현

Assignees

• 주식회사 투바이프롬

Dates

Publication Date

20260513

Application Date

20250121

Claims (6)

1. An integrated management system for AI-based sales optimization and revenue growth, characterized by comprising: a data collection module that collects user purchase history, preferences, budget, visit frequency, product price information, and business partner data in real time and refines and transforms the collected data; a data analysis module that calculates price volatility scores and similar customer scores based on data provided by the data collection module and analyzes user purchase patterns and preferences through clustering; a dashboard module that visually displays the price volatility scores, similar customer scores, and business partner priority scores calculated by the data analysis module and provides warnings for data exceeding a threshold; a recommendation module that recommends customized travel and distribution products for each user based on user budget and visit frequency; and a business partner management module that calculates the sales priority of business partners and designs sales paths based on the location, size, purchase history, and movement overlap of business partners. The above data collection module is, Normalize user-specific purchase history, preferences, budget, visit frequency, and product price data into a range of 0 to 10, and User purchase history data is normalized based on the number of purchases and purchase amount during a specific period (e.g., the last 3 months), and Daily update of price data for airline tickets, hotels, and general retail products, and perform outlier removal and data transformation via JSON, CSV, and SQL, It is characterized by periodically updating and storing data on the location, size, past purchase history, and movement overlap of business partners in a database, and An integrated management system for AI-based sales optimization and revenue growth, characterized by comprising: a data collection module that collects user purchase history, preferences, budget, visit frequency, product price information, and business partner data in real time and refines and transforms the collected data; a data analysis module that calculates price volatility scores and similar customer scores based on data provided by the data collection module and analyzes user purchase patterns and preferences through clustering; a dashboard module that visually displays the price volatility scores, user recommendation scores, and business partner priority scores calculated by the data analysis module and provides warnings for data exceeding a threshold; a recommendation module that recommends customized travel and distribution products for each user based on user budget and visit frequency; and a business partner management module that calculates the sales priority of business partners and designs sales paths based on the location, size, purchase history, and movement overlap of business partners. The above data collection module is, Normalize user-specific purchase history, preferences, budget, visit frequency, and product price data into a range of 0 to 10, and User purchase history data is normalized based on the number of purchases and purchase amount during a specific period (e.g., the last 3 months), and Daily update of price data for airline tickets, hotels, and general retail products, and perform outlier removal and data transformation via JSON, CSV, and SQL, It is characterized by periodically updating and storing data on the location, size, past purchase history, and movement overlap of business partners in a database, and The above data analysis module is, (a) Calculate a price volatility score (Sv) representing the urgency of a purchase decision by combining the standard deviation of price data (pi) over the past n days, the log ratio of the current price (Pcur) to the past price (Pref), and the periodicity (sin(wΔt)) over time (Δt) through multiplication; (b) Calculate a User Recommendation Score (Sr) representing the validity of personalized product recommendations by combining the Euclidean distance between the user's purchase history (ph) and preferences (pf), the similar customer group score (ps), the log-transformed budget (Bu) and visit frequency (Vu), and the periodicity (cos(Φ)) reflecting the promotion period; (c) Characterized by calculating a client sales priority score (Sc) representing the efficiency of a salesperson's visit by combining a log-transformed distance to the client (ld), a square root-transformed client size (qs), a purchase history (gh), and a synergy effect obtained by transforming the difference between the route overlap (do) and the seasonal characteristic (sin(θ)) into an exponential function. The above price volatility score (Sv) is, For the pi data, which is the daily average price over the past 7 or 14 days, the standard deviation term √((1/n)∑(pi-p)^2) represents the range of price fluctuation; A log term log((Pcur/Pref)+1) that reflects the non-linear price change effect by applying a logarithmic function after adding a constant 1 to the ratio of Pref (the average price over the past 30 days or the price 5 days ago) to the current price Pcur to prevent the logarithmic domain from becoming zero; A periodic term sin(wΔt) in which the frequency (w) is set to 2π to reflect a 7-day cycle and the elapsed time (Δt) is input as a unit; and A scaling factor (Kv) set by the administrator to prevent the score from becoming excessively large or small; It is calculated by combining all four of the above terms through multiplication, and It is characterized by classifying products with a calculated Sv value of 1,000 or more as rapidly changing products and products with a value of less than 200 as low volatility products, and The above user recommendation score (Sr) is, A Euclidean combination term √(ph^2+pf^2) obtained by squaring and summing the purchase history score (ph), normalized to a range of 0 to 10 based on the number of purchases and total amount over the past 3 months, and the preference score (pf), normalized to a range of 0 to 10 based on the combination of in-app star ratings, wishes, and survey results, and then taking the square root; A clustering effect term (ps+1) that guarantees a minimum scaling factor of 1x by adding a constant 1 to a similar customer score (ps) set to a range of 0 to 5 based on K-Means or DBSCAN clustering results; A user engagement term formed by combining a budget score (Bu) with a monthly spending limit normalized to a range of 0 to 10 and a visit frequency score (Vu) with the number of accesses in the last month normalized to a range of 0 to 10, converted to a log scale; and cos(Φ), a periodic term designed as a cos function to weight the promotion period at the beginning or end of the month; It is calculated by combining all four of the above terms, and It is characterized by classifying a top recommendation as a target for sending push notifications if the calculated Sr value is 50 or higher, and excluding it from the recommendation target if it is less than 10. The above client sales priority score (Sc) is, A location term ln(ld+1) that non-linearly reflects distance advantage by adding a constant 1 to a location score (ld), which is a distance calculated via a map API converted to a range of 0 to 10, and then taking the natural logarithm (ln); A scale term √(qs) that mitigates the influence of large clients by taking the square root of a scale score (qs) with monthly average sales or order volume set to a range of 0 to 1000; A history term (gh+1) that guarantees the potential of a new business partner by adding a constant 1 to a purchase history score (gh) graded from 0 for new partners to 10 for key partners; and An exponential function term exp(-(do-sin(θ))^2) whose exponential is the square of the difference between the degree of movement overlap (do), assigned a range of 0 to 5 for the efficiency of movement overlap due to visits to adjacent clients, and the seasonal characteristic (sin(θ)), set a range of -1 to +1 for quarterly sales priority, multiplied by a negative constant; It is characterized by being calculated by combining all four of the above terms through multiplication, The above-mentioned customer management module is, Utilizing the principle that the client sales priority score (Sc) is maintained at a high level as the difference between the movement overlap (do) and the season characteristic (sin(θ)) decreases and the value of the exponential function term approaches 1, and the Sc score decreases rapidly as the difference increases and the value of the exponential function term approaches 0; It is characterized by automatically generating a list of clients whose calculated final Sc score exceeds 80 points and proposing them as 'priority visit paths for this week,' and automatically reclassifying clients with a final Sc score of less than 30 points as 'targets for consideration in the next quarter.' The above system is, The above data collection module regularly updates price, user, and business partner information every day; The above data analysis module recalculates the price volatility score (Sv), user recommendation score (Sr), and client sales priority score (Sc) based on the updated information; The above dashboard module visualizes products whose Sv scores exceed a threshold with a 'surge risk' warning, and marks customers whose Sr scores exceed a threshold as 'immediate recommendation'targets; The above-mentioned client management module automatically designs visit routes in order of clients with the highest Sc scores and provides them to the sales team; An AI-based integrated management system for sales optimization and revenue growth, characterized by performing a process of adjusting scale factors (Kv, Kr, Kc) and thresholds by reflecting actual sales and visit results back into the database one day later.
2. delete
3. delete
4. delete
5. delete
6. delete

Description

Integrated Management System for Sales Optimization and Revenue Enhancement Based on Artificial Intelligence The present invention relates to an integrated management system for artificial intelligence-based sales optimization and revenue growth, and is a technology designed to support essential sales activities for improving revenue, particularly for small and medium-sized travel agencies and small and medium-sized distributors. Specifically, this relates to a system that helps increase the success rate of sales by collecting market data such as flight, hotel, and product prices in real time to identify the market competitiveness of a company's products, and by analyzing customer data such as purchasing history, preferences, and budgets to recommend the optimal timing for sales and customized products. In particular, the present invention utilizes collected data to provide real-time data analysis and a visual dashboard through a cloud system. This relates to technology that optimizes data-driven decision-making in the travel sector—such as corporate training, incentive travel, and group events—and in the distribution sector—such as product purchasing, inventory management, and delivery management—thereby supporting efficient sales activities even with limited manpower. Sales activities and client management in travel agencies and the distribution industry have traditionally relied heavily on the individual capabilities and experience of sales representatives. In particular, small and medium-sized enterprises (SMEs) faced limitations in efficient sales decision-making due to the difficulty of systematic data management and analysis caused by limited personnel. Technologies currently on the market have the following problems. First, existing travel agency management systems focus solely on simple reservation management and schedule coordination, making it difficult to provide customized proposals that consider the characteristics and preferences of client companies. In particular, when organizing group events such as corporate training or incentive travel, there were difficulties in securing price and product competitiveness due to the lack of features to recommend optimal products within a budget or to predict the optimal time for purchase by tracking real-time fluctuations in flight and hotel prices. Second, current distribution management systems focus solely on inventory and delivery management, failing to provide strategic insights for sales activities. It lacks the functionality to support proactive sales activities by predicting product price volatility or analyzing client purchasing patterns. In particular, the functionality to generate optimized proposals for bulk or recurring purchases targeting corporate customers is lacking. Third, current client management systems remain at the level of simple customer information databases and fail to support the establishment of efficient sales activity plans. Advanced features are lacking, such as designing optimal sales channels that comprehensively consider the location, size, and purchase history of clients, or recommending customized products for each client. In particular, it lacks the function to identify new sales opportunities by analyzing the relationships between similar client companies. Against this backdrop, there is a need to develop a new system that analyzes data in the travel and distribution sectors and utilizes artificial intelligence technology to support efficient sales activities. In particular, there is a need to develop an intuitive cloud-based system that takes into account the realistic operating environment of small and medium-sized enterprises (SMEs) operating with limited manpower due to the difficult economic climate. Figure 1 illustrates an overall relationship diagram according to the present invention. FIGS. 2 to 5 schematically illustrate an embodiment of the present invention. Hereinafter, various embodiments are described in more detail with reference to the attached drawings. The embodiments described in this specification may be modified in various ways. Specific embodiments may be depicted in the drawings and described in detail in the detailed description. However, specific embodiments disclosed in the attached drawings are intended only to facilitate understanding of various embodiments. Accordingly, the technical concept is not limited by specific embodiments disclosed in the attached drawings, and it should be understood that it includes all equivalents or substitutions that fall within the spirit and scope of the invention. Terms including ordinal numbers, such as first, second, etc., may be used to describe various components, but these components are not limited by the aforementioned terms. The aforementioned terms are used solely for the purpose of distinguishing one component from another. Functions related to artificial intelligence according to the present disclosure are operated through a processor and memory