BACKGROUND
Technical Field [001] The embodiments herein relate to customer relationship management and, more particularly, to customer loyalty programs in a retail environment.
Description of the Related Art
[002] Currently available customer shopping interactive systems are limited to providing pricing and product identification information to the customers. The customers are provided discounts without any personalization. A brand provides discount if it identifies that the customer purchased a competing brand. In addition, the loyalty programs are normally "fit for all" schemes. So, avail "20% discount" or "buy one get one free" etc is the messages we see. The second aspect is offering it to the customer. Some latest developments suggest using RFID based position identification and product identification techniques to advertise the loyalty programs as the customer is in the aisle. The conceptual problems associated with the methods are that the discount/sale/advertisement is not a great stimulator as there exists non-technical alternatives. Also all these techniques only refer to pre-defined discounts. What provides greatest flexibility is if loyalty plans is constructed on the fly for each customer distinctly and integrate that seamlessly with the POS system.
[003] In case of RFID techniques providing information to the customers there are operational problems associated with the same as RFID is cost ineffective to be used on low-cost retail items and requires a fundamental shift in the way retail stores do their

sales. There are many technical challenges associated with deployment of RFIDs. For example, there are problems with false or missing reads as a result of radio waves being easily distorted, detected, absorbed, or interfered with. There are a number of system-level challenges such as determining the number, type and placement of readers. For any customized recommendations it is extremely important to identify the customer. The loyalty cards which can be shared between people can lead to a wrong set of recommendations negating the whole objective of real-time discounts. For implementing such real-time discounts the biggest challenge is in fact integration of discounts offered with the POS systems. For most retailers this integration is what is working against using innovations. Some of the technique combines this with self-checkout. They fail because of pilferage. So far, discounts/loyalty programs are treated as predefined business rules and hence fail to provide a cost effective solution.
SUMMARY
[004] In view of the foregoing, an embodiment herein provides a system and method of achieving real time discount plans to simulate cart swell in a consumer interactive shopping system.
[005]Embodiments further disclose a system and method of providing real time discounts to the customers. The device is hand held or can be attached to any shopping chart. An algorithm for the method is included. A customer becomes a member by entering his details on a computer at the store. The computer is provided with a camera and bio-metric authenticator. As soon as the registration is done, the server notes the details of the customer. Registration can be done from the internet or individual hardware

equipment. The device system has software with three different functionalities. It allows the user to manually enter a product he/she wants to buy and displays that product and other products he/she is most likely to buy along with their discounts. He will also be able to scan the item using the scanner (one of the attachments to the handheld device). Generates a bar code with a specific price (through an attached printer). And also identifies a customer via bio-metric using a finger print scanner. The customer enters the product required in the handheld device or scans it. This detail is sent to the central computer. The computer performs an advanced non-linear recommendation analysis using an optimization algorithm, the central server decides on a discount. As the customer picks a new product, the analysis and hence the products in the hand-held get updated. Using a complex behavioral model, the system computes all the items that this customer might like and an appropriate discount the customer is eligible for and displays them. Each time a customer buys and scans the product through bar-code scanner the algorithm updates the model. When the customer completes his/her purchases, they go to the POS and handover their hardware to the sales agent. The agent scans the items and the hardware prints a discount bar code which is scanned by the sales person to implement the discount on the custom items.
[006] These and other aspects of the embodiments herein will be better appreciated and understood when considered in conjunction with the following description and the accompanying drawings.

BRIEF DESCRIPTION OF THE DRAWINGS
[007] The embodiments herein will be better understood from the following detailed description with reference to the drawings, in which:
[008]FIG. 1 illustrates the diagram of the system elements to stimulate cart swell, in accordance with the embodiments herein;
[009]FIG. 2 illustrates the various data inputs for the statistical optimization algorithm, in accordance with the embodiments herein;
[0010] FIG. 3 is a flow chart depicting the process flow in statistical optimization algorithm, in accordance with the embodiments herein;
DETAILED DESCRIPTION OF EMBODIMENTS
[0011] The embodiments herein and the various features and advantageous details thereof are explained more fully with reference to the non-limiting embodiments that are illustrated in the accompanying drawings and detailed in the following description. Descriptions of well-known components and processing techniques are omitted so as to not unnecessarily obscure the embodiments herein. The examples used herein are intended merely to facilitate an understanding of ways in which the embodiments herein may be practiced and to further enable those of skill in the art to practice the embodiments herein. Accordingly, the examples should not be construed as limiting the scope of the embodiments herein.
[0012]The embodiments herein achieve a system and method for real time discount plans to simulate cart swell in consumer interactive shopping system by incorporating hardware and an algorithm for the process flow. Referring now to the

drawings, and more particularly to FIGS. 1 through 3, where similar reference characters denote corresponding features consistently throughout the figures, there are shown embodiments.
[0013] A system and method of achieving real time discount plans to simulate cart swell in a consumer interactive shopping system is disclosed. The device can be hand held and can be attached to a shopping chart. An algorithm for the method is included. A customer becomes a member by entering his details on a computer at the store. The computer is provided with a camera and bio-metric authenticator. The person gets a loyalty card if he prefers. Loyalty card is not a must as the entire system is associated with a bio-metric identifier. As soon as the registration is done, the server notes the details of the customer. Registration can be done from the internet or individual hardware equipment. The system has software with three different functionalities. System allows the user to manually enter a product he/she wants to buy and displays the product and other products he/she is most likely to buy along with their discounts. The user can scan the items he bought also. The system generates a bar code with a specific price. And also identifies a customer via bio-metric scanning. The customer enters the product required in the device by entering in the designated number or by scanning the product. The details are sent to the central computer. The computer performs an advanced non-linear recommendation analysis using an optimization algorithm, the central server then decides on a discount. The data is sent to the hand-held. As the customer picks a new product, the analysis and hence the products in the hand-held get updated. Using a complex behavioral model, the system computes all the items that this customer might like and an appropriate discount the customer is eligible for and displays them. Each time a customer

buys and scans the product through bar-code scanner the algorithm updates the model. When the customer completes his/her purchases, they go to the POS and handover their device to the sales agent. The agent scans the items and the device prints a discount bar code which is scanned by the sales person to implement the discount on the custom items. The system motivates the customer to stroll more across the store, buy more and enjoy real time discounts.
[0014]FIG. 1 illustrates the diagram of the system elements to stimulate cart swell, in accordance with the embodiments herein. The device 101 comprises of a handheld device 105, a Biometric authentication device 102, TLP printer 103 and a barcode scanner 104. The biometric authentication device 102 acts as a finger print scanner for authencation of the customers. The TLP printer 103 prints the barcodes of the products. A customer becomes a member by entering the details in a kiosk placed near a store. The kiosk will be a computer with touch screen monitor with a camera and bio¬metric authenticator 102. The person gets a loyalty card if he prefers. Loyalty card is not a must as the entire system is associated with a bio-metric identifier. As soon as the registration is done, the server notes the details of the customer. Registration can be done from the internet or individual hardware equipment. The device 101 has software with three different functionalities. At first the device 101 allows the user to manually enter a product he/she wants to buy and displays that product and other products he/she is most likely to buy along with their discounts. Then, the device 101 generates a bar code with a specific price. And identifies a customer via the bio-metric identification device 102. When a customer decides to buy some thing. He enters his request in to the handheld device 105, a web site or SMSes to a designated number. The handheld device 105 may

be a Personal Digital Assistant (PDA) or any other handheld device capable of accepting an input from the customer and communication through wireless means with the server. So, the request of items can be done even before reaching the shop. The details are sent to the central computer. Now, the computer performs an advanced non-linear recommendation analysis and identifies the products that customer is most likely to buy. Using an optimization algorithm, the central server decides on a discount. The optimization engine takes care of the transaction score, product score, customer score and floor manager's view to decide a personalize discount. The data is then sent to the device 101. As the customer picks a new product, the analysis data and the products in the device 101 get updated. There is a built-in navigation detail for the products recommended. As soon as the customer picks the item, they scan it using the bar code. Using a complex behavioral model, the server computes all the items that the customer might like and an appropriate discount the customer is eligible for and displays them. Each time a customer buys and scans the product through bar-code scanner the algorithm updates the model. As the customer continues to shop and the data gets updated at real-time. When the customer completes his/her purchases, they go to the POS and handover their device 101 to the sales agent. The agent scans the items (once on her scanner and once on the device's scanner). At the end of scanning, the device 101 prints a discount barcode which is scanned by the sales person to implement the discount on the custom items. In addition the server is provided with a statistical algorithm for the optimization. Initially each of the customers will be clustered to identify like customers based on the demographic profile. In each of the cluster, based on the past purchase data of individual customer the similar customer, frequency of individual customer visiting the store,

amount he/she is buying, last visited, discount friendliness of each customer will be scored in the final equation. This server will capture manager's input on each product depending on the competitor's promotion or any other extraneous market situation. This information will also be used for improvement of the optimization model performance. One more interesting and novel way to capture buying interest of customer is to identify the like customer in real-time. When a customer in a particular store is actually buying (real-time), based on their purchases a mathematical algorithm will identify the similar customers on the store real-time and then invite them to interact with the other customers. They have to swap each others card to let the system know that they are interested in a combined buying.
[0015] FIG. 2 illustrates the various data inputs for the statistical optimization algorithm, in accordance with the embodiments herein. The various inputs for the optimization algorithm 201 are customer score 206, product score 205, transaction score 203, referral score 204 and manager score 202. Initially each of the customers will be clustered to identify like customers based on the demographic profile. In each of the cluster, based on the past purchase data of individual customer the similar customer, frequency of individual customer visiting the store, amount he/she is buying, last visited, discount friendliness each customer will be scored. So the final equation is as shown below.
Y = f(r,fr,m,d).
[0016]Where Y is customer score which is a function of recency (r), frequency (fr), Monetary Value (m) and discount friendliness (d). The overall score Y will be used in the final optimization equation as an independent variable to finally decide the loyalty.

Similarly each product on the store will be scored based on transaction volume, frequency of stock-outs, shelf life, replenishment, margin, seasonality, supervisor input etc.
E = f(v,st,sl, rep,mr,se,su)
[0017] Where E is product score which is a function of transaction volume (v), frequency of stock outs (st), shelf life (si), replenishment (rep), margin (mr), seasonality (se), supervisor input (su). This variable is another independent variable for the optimization. On any given visit to the store each transaction will be scored giving weights to the amount bought. Moreover the entire database will be mined for establishing the product affinity relationship. The product affinity relationships help in identifying the similar purchase behavior. On any particular transaction this statistics (T) will be calculated based on individual purchases. It will get computed on each product being purchased on real-time. The system will capture manager's input (converted to consolidated numeric variable M) on each product depending on the competitor's promotion or any other extraneous market situation. This information will also be used for improvement of the optimization model performance.
[0018] One more interesting and novel way to capture buying interest of customer is to identify the like customer in real-time. When a customer in a particular store is actually buying (real-time), based on his purchases a mathematical algorithm will identify the similar customers on the store in real-time and then invite them to interact with the other customer. The customers have to swap each others card to let the system know that they are interested in a combine buying. As an example person A and B are buying now and based on their purchases the algorithm identifies that they are similar customer. It

will invite each of the customers to interact. If they agree and swap their cards, then the system will start keeping track of what they are buying after they became friends. As the system already knows what the customers have individually bought, if A buys some product after becoming friend with B, which B has bought already then B will get some discount assuming that A has bought those product because of the referral from B. This will help in building a network of buyers and in turn more cross-sell. This referral score (|i) for each of the customers will be captured by the proposed system and it will be used in the final optimization model for final discount calculation. All this data will be used to run an optimization algorithm to arrive at a better loyalty program for each individual customer. In addition the mathematical model predicts the product shelf life, customer value, product margin to arrive at a customized discount to the customer.
[0019] FIG. 3 is a flow chart depicting the process flow in statistical optimization algorithm, in accordance with the embodiments herein. The device 101 can be handheld or can be attached to a shopping cart. A customer becomes a member by entering his details on a computer at the store. The computer is provided with a camera and bio-metric authenticator. As soon as the registration is done, the server notes the details of the customer. Registration can be done from the internet or individual hardware equipment. The customer picks up the product of his choice from the store. Entry of the product is made (301) on the device 101. The customer enters the product required in the device by entering in the designated number or by scanning the product. The authentication and registration of the product is done (302). The details of the product to be purchased are provided (303) on a hand held device such as a PDA, a website or the like. The information of the product is sent (304) to the barcode scanner 104. The barcode scanner

sends the product information through (305) the WiFi to a central server. The centralized server is fed with a statistical optimization algorithm that performs real time personalized discounting. Various data inputs of the algorithm are fed to the centralized server such as customer score 206, product score 205, transaction score 203, referral score 204 and manager score 202. Initially each of the customers will be clustered to identify like customers based on the demographic profile. In each of the cluster, based on the past purchase data of individual customer the similar customer, frequency of individual customer visiting the store, amount he/she is buying, last visited, discount friendliness each customer will be scored. The server computes (306) the items that need to be given discount, the amount of discount, a discount code and image of that item. Further the discounted product list along with the amount information which can be utilized by the customer for biding and purpose is sent (307) to display on the monitor.
[0020]The process repeats for every new item. As the items are scanned, the system knows how many discounted items are added to the cart. The server makes a check (308) as to whether the customer has finished shopping. This is accomplished ay the algorithm that keeps tracking of the entry of any new items. In case there is no more entry of the items the shopping is assumed to be completed. In case the shopping is not completed the entry of the new item is passed to the barcode scanner 104. On completion of shopping a barcode of discounts offered for the purchase is generated (309). The receipt of the discount is printed (310) by the TLP printer 103. Before check out at the POS the products are scanned twice and the discount is implemented (311) at the POS by scanning the discount code generated. The product is (312) further checked out.