FORM 2
THE PATENTS ACT, 1970
(39 of 1970)
&
THE PATENT RULES, 2003
COMPLETE SPECIFICATION
(See Section 10 and Rule 13)
Title of invention:
A SYSTEM AND METHOD TO SIMULATE PRODUCT SHARE AT DIFFERENT ATTRIBUTES AND LEVELS
Applicant:
TATA Consultancy Services Limited
A company Incorporated in India under The Companies Act, 1956
Having address:
Nirmal Building. 9th Floor,
Nariman Point, Mumbai 400021.
MaharashtraT India
The following specification describes the invention and the manner in which it is to be performed.

FIELD OF THE INVENTION
The present invention relates to a field of conjoint analysis. More specifically the invention relates to the field of simulating a new product at different price points and different attributes by analyzing the existing products in the market.
BACKGROUND OF THE INVENTION
With the development in the industrial sector, the frequency of new products introduction in the market has drastically increased. Corporates, in order to stay in competition need to analyze the market and take appropriate decisions before launching a new product and discontinuing the old products. It is difficult for a brand manager to know the probable market share after introducing a product at a particular price point with particular attributes. Without knowing the probable performance of product to be launched in the future and kind of response it may achieve in the market, it is difficult to derive an assembly line for the manufacturing of the product. This can also affect the demand supply chain of a particular product and thus may lead to unnecessary investments.
While designing a new product, the brand manager needs to have a complete overview of the current market. The brand manager needs to create different market scenarios with change in price points of own products or competitor products and know the share gain or loss of his new product. Before discontinuing a particular product the brand manager needs to understand the percentage of share the product contributes by comparing within different brands.
The failure or success of a new product largely depends on the precision of market analysis conducted before launching of the product. A detailed study needs to be

done before finalizing the attributes and levels that may help in depicting the failure success ratio of the product to be launched. This type of market analysis is called as conjoint analysis. In conjoint analysis the market share of existing products is analyzed and the different attributes and levels such as packet size, shape, color, package, pricing and pack-sizes is analyzed. In conjoint analysis, market survey is conducted in order to analyze the existing products and derive attributes and levels for the new product to be launched.
But with the increase in the number of products and the attributes and levels at which they are available it becomes difficult to perform conjoint analysis and derive the part worth utility of each attribute at different price points. Few current simulators are built on the aggregated level models. These simulators lack effective level-wise calibration of product attributes to simulate the performance of the product to be launched and hence lack desired data to forecast future market share of the product in competence with the similar products. Further, in the present scenario, the simulating tools available in the market lacks feasibility of refining the product attributes and is limited to focus on the historical data obtained from market surveys and does not take account the instantaneous response from the users pertaining to attributes and the desired levels for determining future market share. This may result in forecasting incorrect or partially correct market share of the product to be launched. Additionally, the existing tools lack prediction of future performance of the product by considering parameters such as elasticity and profitability of a new product at the time of design of a new product.
Hence in view of the above lacunae in the art, there is a long-felt need to derive a simulation technique that is configured to cumulatively perform individual level attributes as well as aggregated level attributes analysis while determining the market share of a product to be launched. Further, there is a need for a simulator built on the

hierarchical model output capable of generating different market scenarios by calibrating the attributes with respect to different levels for simulating the performance of the product, wherein the attributes and levels involved in calibration process includes both the historical attributes and levels as-well-as instantaneous attributes and levels captured from the users related to future product portfolio. Also there is a need to determine parameters like elasticity and profitability of a new product at the time of designing a new product. Further, there is need for interactive dashboard facilitating the user to derive insights depicting market share of the product to be launched by means of displayed summary and analysis by the conjoint scenario analysis to arrive at a conclusion with the best possible combination and scenario.
OBJECTS OF THE INVENTION
The primary object of a present invention is to provide a system and method to enable simulator tool capable of simulating different market scenarios for predicting the market share of a new product to be launched.
Another object of the invention is to provide a system and method that enables efficient prediction of market share of a particular product using a limited set of historical data.
Yet another object of the present invention is to provide a system and method that provides an interactive dashboard displaying the results of calibration oriented scenario analysis to arrive at a conclusion with the best possible combination and scenario.

Yet another object of the invention is to provide a system and method that enables multi-level attribute-based scenario analysis by referring a plurality of libraries storing historical information and thereby facilitating data reusability.
Yet another object of the invention is to provide a system and method that enables instantaneous capturing of new set of attributes and levels unavailable through historical data from the consumers on future product portfolios.
Yet another object of the invention is to predict the part worth utilities for the attributes and levels obtained from both historical data and those captured from the users.
Still another object of the invention is to enable a system and method which is built on the hierarchical model output where user has flexibility to generate different market scenarios and visualize the results to determine the best possible scenario.
SUMMARY OF THE INVENTION
Before the present systems and methods, enablement are described, it is to be understood that this application is not limited to the particular systems, and methodologies described herein, as there can be multiple possible embodiments which are not expressly illustrated in the present disclosures. It is also to be understood that the terminology used in the description is for the purpose of describing the particular versions or embodiments only, and is not intended to limit the scope of the present application.
In one embodiment, the present invention discloses a system and method to simulate the future performance of a new product. For this purpose different attributes such as size of product, color of the packet, flavor and product promotion are considered as

primary attributes. Apart from these attributes other product specific attributes like megapixel of camera, resolution, operation system for mobile, any attributes for mobile operator, formulary status for pharmacy, auto manufacturing segment and type of packaging etc. associated with the existing products are determined. The levels such as different price points, and packet-sizes, different kind of promotions (e.g. 5%off or 10%OTT), different megapixel of mobiles and different operating systems etc. at which the existing products are available is dynamically modeled in order to generate various simulation scenarios. Historical data associated with one or more existing products is analyzed to derive a set of attributes and levels for scenario analysis. The system is further adapted to derive a new set of probable attributes and it various level considering the future product portfolio similar to said product to be launched.
In one embodiment, a profile is derived for said new product to be launched based upon the identified and derived attributes and levels. A choice card is used in order to determine the preferences of multiple users related to existing products at said attributes and levels. This enables selection of a sub-set of attributes and levels in context with the designed profile of a new product. A fractional factorial design is used that enables to reduce the number of scenarios and correctly estimate the part worth utility of each product. An analysis is performed using the Hierarchical Bayesian model on the selected sub-set of attributes and levels to determine the part worth utilities of each selected sub-set of attribute at different levels. Finally, a simulator is configured to simulate plurality of scenarios by calibrating the selected attributes at each level and their associated part-worth utilities. This simulator determines the best possible scenario and the performance of the product in these scenarios that enables forecasting future market share of the product.

BRIEF DESCRIPTION OF DRAWINGS
The foregoing summary, as well as the following detailed description of embodiments, is better understood when read in conjunction with the appended drawings. For the purpose of illustrating the invention, there is shown in the present document example constructions of the invention: however, the invention is not limited to the specific methods and apparatus disclosed in the document and the drawings:
Figure I is a block diagram illustrating the various computing devices connected over a communication network involved in the process of historical data gathering and analysis for launching a new product.
Figure 2 is a system architecture diagram illustrating the different modules and hardware units collaborated together in order to simulate market share of a new product.
Figure 3 is a flow diagram illustrating different steps involved in the process of simulating a product at different attributes and levels.
DETAILED DESCRIPTION
The description has been presented with reference to an exemplary embodiment of the invention. Persons skilled in the art and technology to which this invention pertains will appreciate that alterations and changes in the described method and system of operation can be practiced without meaningfully departing from the principle spirit and scope of this invention.

Jn accordance to one embodiment of the present invention, a system and method to simulate a new product at different attributes and levels is disclosed. The system work on individual level modeling and generates market share within any market scenario by analyzing historical data associated with the existing products and extracting a historical set of attributes and levels. These historical set of attributes and levels is updated with a new set of attributes and its associated levels accepted from a user in virtue of the future products and an updated set of attributes and levels is generated. Further the system predicts changes in variable entities like market share, elasticity, market revenue, market volume, etc. for this purpose the simulator take the part-worth utility of each of the updated attribute at different levels in order to derive the final market share of the new product. The simulator also considers attributes such as distribution and availability of the goods for the prediction purpose. Thus the user has the flexibility to change both these attributes at the time of simulation.
In one embodiment the simulator predicts the part worth utilities of updated set of attributes and levels by making use of hierarchical Bayesian technique. Further the simulator enables segment level market share generation for all products. For this purpose a calibration factor is used to match the survey based share with the current market share of the products. This calibration factor is considered to adjust the sample mean and variance with population mean and variance in order to determine the distribution of that product. The distribution is the percentage of market availability of a particular product corresponding to total market population. The simulator has flexibility to take care of private level total share at the time of projection of the final share.
Referring to figure 1 is a block diagram illustrating the various computing units in a system (100) coupled together using a communication network (107) for historical

data gathering and analysis. The system (100) comprises of a plurality of POS machines (101), a Central Server (103) coupled to a central repository (105) storing a historic data (111), a client machine (109) with data transfer capabilities and a communication network (107) enabling communication amongst the client machine (109), Central Server (103) and the POS machines (101).
In one embodiment, the raw historical data from a plurality of POS machines (101) is captured over a period of time and sent to a Central Server (103) using the communication network (107). The Central Server (103) processes the raw historical data from multiple sources and stores it into the central repository (105) as segregated historical data (111). The processing of the raw historical data takes place at various stages in order to classify it into various historical attributes and levels associated with each product. The Central Server (103) further analyzes the segregated historical data (111) and generates a base case scenario for new product simulation. Input is accepted from the client machine (109) regarding the introduction of new attributes and the variation in the level associated with the new product. These are set of attributes for which there is no historical data available in the central repository (105). Therefore, these attributes and levels are instantaneously captured from the client machine (109) focusing on future products similar to said product to be launched and their performance constraints. This information is sent to the Central Server (103) which then combines the new attributes and the historical attributes to generate an update set of attributes and their associated levels. The new product is modeled at various updated attributes and levels.
Referring to Figure 2 is a system architecture diagram illustrating the different modules and hardware units arranged together in order to simulate market share of a new product in an embodiment of the present invention. The Central Server (103) comprises of Attributes and Levels determination module (201), a design of

experiments module (203), a choice model (205) and a simulator (207). The Central Server (103) is further coupled to a client machine (109) which further comprises of an input module (209), a memory unit (211), a display unit (213) and a dashboard (215). The central repository (105) storing the segregated historical data (111) is coupled to the Central Server (103).
In one embodiment of the present invention, the segregated historical data (111) in the central repository (105) is analyzed by attributes and Levels Determination module (201) to extract the attributes such as the shape, color and packaging of the product etc. Further, the attributes and Levels Determination module (201) is configured to extract the different levels such as price points, pack-sizes, etc. associated with the existing products in the market on which the extracted attributes can be calibrated to generate plurality of scenarios. The attributes and Levels Determination module (201) is initially configured to generate a base case scenario on the basis of which the successive scenarios will be generated by means of variance of different determined attributes on different levels to finally derive a best-case scenario depicting the performance of the product which is to be launched.
In one embodiment of the present invention, the attributes and Levels Determination module (201) also enables considering of the new attributes and levels specified by a user related to products not available in the existing market. A need state analysis is performed in order to analyze probable attributes from the consumer. After the system identifies a need for additional attributes on the basis need-state analysis, , these additional attributes are then included into the choice cards. In order to enable this, the input module (209) on the client machine (109) is configured to accept the user input in the form of new attributes and levels. This input is transmitted to the Central Server (103) using the communication network (107). In this embodiment, the design of experiment module (203) enables selection of a sub-set of attributes

and levels in order to generate a design profile for the new product. The Design of Experiment module (203) is configured to work on the principle of choice-based modeling, wherein a choice card is made available for the user to select a sub-set of attributes and levels from the updated attributes and levels. The Choice modeling is performed in order to capture the interest of stakeholders that is correlated with the profile design of the new product.
In one embodiment of the present invention, the output of choice-card is accepted by a Hierarchical Bayesian analysis module (205) is configured to perform Hierarchical Bayesian model analysis on the selected sub-set of attributes and levels that result in determining of the part worth utility for each of the selected sub-set of attribute at different levels. These part worth utilities are used as an input to the simulator (207). The simulator (207) first finalizes the design and then applies different statistical analysis techniques for predicting the future market share of the new product by considering the part-worth utilities and availability of the product. For example, in an embodiment, the simulator (207) intelligently enables calibration of product attributes at different levels to obtain a market share value at each level. A plurality of scenario analysis is applied using multi-level product attributes to finally settle with the best combination of current market share, future market share and the corresponding attribute to be adopted for the obtained desired share values. The dashboard (215) is configured to generate various analytical information and statistical graphs deriving insights of the simulation analysis. Such interactive summary in congruent with the variance of attributes corresponding to variance in levels enables adjudging the best case scenario for the user that may result in positive impact for product marketing. The statistical graphs are displayed on the display unit (109) of the client machine (213).

Referring to figure 3 is a flow diagram illustrating different steps involved in the process of simulating a product at different attributes and levels. The process of simulation starts at step (301) wherein the number of existing products to be analyzed is decided. The price points at which these products are to be analyzed is also decided in this step. For this purpose user input is accepted using a client machine regarding the number of products and their updated attributes and levels be considered. At step (303) the number of combinations required to address all possible scenarios is determined by the Central Server. At step (305) a Design of experiment is executed on the design profile to determine the minimum number of scenarios required to simulate the exact market share of the new product. This is done by applying a fractional factorial design technique in order to reduce the number of combinations without affecting the accuracy of the prediction. At step (307) different market surveys are conducted in order to gain inputs from user community regarding their likes and dislikes for a particular product with certain attributes at a particular price point. The inputs received from various surveys conducted is analyzed using a hierarchical Bayesian technique to determine the part worth utility of each attribute at the specified price point at step (309). At step (311) inputs such as current market share of the products and the availability of the goods in the market are used by the simulator in order to predict the market share of the new product under simulation. At step (313) the survey based share is compared with the present share of the product. The survey mean and standard deviation is different than population mean and standard deviation. The system matches sample mean and standard deviation from the DOE output to the population means and standard deviation. This test represents the accuracy of the simulator in predicting the market share of the product. Finally at step (315) all the scenarios identified at step (303) are simulated using the simulator.

WORKING EXAMPLE
The process of simulation of a new product is divided into five different stages. These are as follows:
1. Defining different market scenarios
2. Applying design of experiment to reduce the number of choices
3. Performing market survey on the generate scenarios using choice cards
4. Applying Hierarchical Bayesian model to predict the part worth utility
5. Using simulation technique to generate all scenarios
Analysis is performed at each of these stages to determine the part worth utility of each attribute.
In an exemplary embodiment, consider a scenario for calculating a market share of a cosmetic product that needs to be launched. The present invention enables generation of a plurality of scenarios for determining the future market share. In this exemplary embodiment, consider the simulation analysis being conducted on eight similar cosmetic products having variable packet size attribute including SKU1 350mL, SKU2 473mL, SKU3 500mL, SKU4 591mL, SKU5 750mL, SKU6 473mL; SKU7 475mL, SKU8 750mL. The simulation process is applied on these variable cosmetic products in different stages to finally obtain the market share of the new cosmetic product to be launched which as explained as below:
1. Defining different market scenarios:-
In this exemplary embodiment, to illustrate the actual simulation, only one scenario is considered wherein only price is varied with respect to packet size attribute and assuming that all the other attributes are held constant at their current levels. For this scenario, Table 1 appended below illustrates the possible combinations of price that

may be quoted against variable packet size attribute for five different levels at which the price level is varied.

SKU
Name Pack Size Price 1 Price 2 Price 3 Price 4 Price 5
SKUl
350mL 350 1.50 1.60 1.70 1.80 2.00
SKU2 473mL 473 2.00 2.10 2.20 2.30 2.50
SKU3 500mL 500 2.00 2.10 2.30 2.50 2.70
SKU4 591mL 591 3.00 3.10 3.20 3.30 3.50
SKU5 750mL 750 3.00 3.10 3.30 3.50 4.00
SKU6 473mL 473 1.50 1.60 1.70 1.80 2.00
SKU7 475mL 475 1.50 1.60 1.70 1.80 2.00
SKU8 750mL 750 2.50 2.60 2.70 2.80 3.00
All Other 475
Table-I
2. Applying design of experiment to reduce the number of choices:-
In this exemplary embodiment, following the definition of pricing scenarios, the next step is to accept the user preferences for possible combinations determined by means of full factorial consideration principle for the defined scenarios, which would be 58 - 390,625 different combinations. However, in accordance with the present invention, this number of combinations is reduced to 180 combinations by applying fractional factorial technique using orthogonal array concept.
In this exemplary embodiment, the fractional factorial technique applied preserves only those entries which can create an impact over the prediction of part worth of

each attribute while eliminating the other redundant entries. This results in generation of choice card which is illustrated in the table-Il:

Product Price skul Price sku2 Price
sku3 Price
sku4 Price sku5 Price sku6 Price
sku7 Price sku8
SKU1 1 0 0 0 0 0 0 0
SKU3 0 0 5 0 0 0 0 0
Choice SKU5 0 0 0 0 4 0 0 0
Card 1 SKU6 0 0 0 0 0 3 0 0
SKU7 0 0 0 0 0 0 1 0
SKU8 0 0 0 0 0 0 0 4


SKU1 3 0 0 0 0 0 0 0
Choice
Card
180 SKU2 0 2 0 0 0 0 0 0

SKU5 0 0 0 0 4 0 0 0

SKU6 0 0 0 0 0 1 0 0
SKU7 0 0 0 0 0 0 5 0
SKU8 0 0 0 0 0 0 0 5
Table-II
3. Performing market survey on the generate scenarios using choice cards:-
In this exemplary embodiment, the scenarios depicted in the choice card are displayed to respondents one at a time. The choice card is used to capture inputs from the respondents in the form of'+1' and M', respectively as per their choice. In a similar manner multiple scenarios are generated for different attributes and inputs from respondents for the multiple scenarios are captured. New Probable attributes such as package size are also identified in this step by performing need state analysis which is however optional and is dependent on the user preference as per the new product requirements and hence not mandatory. At the end, all the choice cards are fed for conducting Hierarchical Bayesian model analysis in the later stage.

4. Applying Hierarchical Bayesian model to predict the part worth utility:-
In this exemplary embodiment, the choice cards are the fed into the system and Hierarchical Bayesian analysis is performed on each of the attributes. The output of the hierarchical Bayesian analysis is the part worth utility determined of each attributes extracted from the choice card. For this example of cosmetic products, the part worth utilities generated by hierarchical Bayesian analysis are illustrated in the Table-JII Appended below:

SKU Name Initial Pred. Share Adjustment Factor Pred. Share
SKU1 350mL 0
SKU2 473mL 25.1567 1.1057342 29.7726
SKU3 500mL 0
SKU4 59JmL 16.1553 0.2873795 6.9741
SKU5 750mL 0
SKU6 473mL 14.6871 0.9541241 17.7362
SKU7 475mL 39.7896 1.0032643 40.6419
SKU8 750mL 0
All Other 4.2113 3.5213743 4.8752
Table-Ill
In an embodiment, the 'Adjustment Factor' is utilized to perform further study on each sample. The results from the study never match with population and industry level results. The e Adjustment Factor helps to match sample results with population results. This is a calibration factor calculated from sample, industry and population. This factor is same for a particular study but may be varied across different studies.

5. Using simulation technique to generate all scenarios:-
In this step, all the part worth utilities associated with the attributes is then fed into a simulator. The simulator is adapted to determine the impact of each part worth utility and accordingly simulates the market share for the new product to be launched. The output of the simulator is analytical graphs depicting the future market share of the new product, the change in market share based upon the variations introduced by the user with respect to each attribute and combinations thereof.
ADVANTAGES OF THE INVENTION
The present invention has following advantages:
The present invention enables prediction of market share of a new product to be launched by analyzing the existing products at different attributes and levels.
The present invention enables a simulator which can accurately predict the part worth utility of an attribute based upon the limited historical data.
The present invention enables introducing of new attributes and levels associated with the new product and performing analysis to determine part worth utilities of each new attribute at different levels.
The present invention further enables using hierarchical Bayesian model in combination with fractional factorial design to correctly identify the part worth utility of a product using a limited number of combinations of levels and attributes.
The methodology and techniques described with respect to the exemplary embodiments can be performed using a computer-implemented system or other computing device within which a set of instructions, when executed, may cause the

said computer-implemented system to perform any one or more of the methodologies discussed above. The said computer-implemented system may include a processor embedded within the said computer-implemented system which is configured for executing the said programmed instructions or the said set of instructions. The said computer-implemented system is configured from different modules; each module is configured for executing programmed instructions or set of instruction to perform a particular task. According to the embodiments of the present invention, the computer-implemented system may also operate as a standalone device.
In accordance with various embodiments of the present disclosure, the methods described herein are intended for operation as software programs running on a computer processor - [processor embedded within the said computer-implemented system].
Although the invention has been described in terms of specific embodiments and applications, persons skilled in the art can, in light of this teaching, generate additional embodiments without exceeding the scope or departing from the spirit of the invention described herein.

CLAIMS:
1. A method to simulate the future performance of a new product characterized by dynamically modeling the product at different simulating attributes and levels to generate various simulating scenarios thereof, the method comprising process implemented steps of:
a. Retrieving historical data associated with one or more existing
products to determine a historical set of attributes and levels for
scenario analysis;
b. Refining said historical set of attributes and levels by capturing a
new set of attributes and its associated levels from a user for
future products to generate updated set of attributes and levels
therefor;
c. designing a profile for said product and a choice card thereof,
wherein said choice card enables selection of a sub-set of all
updated attributes and levels in context with said designed profile;
d. Performing a Hierarchical Bayesian model analysis on the
selected sub-set of updated attributes and levels to determine the
part worth utilities of each selected sub-set of attribute at different
levels; and
e. Building a simulator configured to simulate plurality of scenarios
by calibrating the selected attributes at different selected levels
and considering part-worth utilities thereof to determine the best
possible scenario depicting the performance of the product.
2. The method of claim 1, wherein the historical attributes are selected from group consisting of shape of the product, color of the packet, type of packaging and combination thereof

3. The method of claim 1, wherein the historical levels of said products are selected from group consisting of different price points, pack-sizes and combination thereof.
4. The method of claim 1, wherein said choice card is designed to determine the preferences of multiple users related to existing products at said updated attributes and levels.
5. The method of claim 1, said design of profile and choice card is enabled by means of fractional factorial design that enables to reduce the number of scenarios and correctly estimate the part worth utility of each product.
6. A system to simulate the future performance of a new product characterized by dynamically modeling the product at different simulating attributes and levels to generate various simulating scenarios thereof, the system comprising: a client machine, a memory; a central server coupled to a central repository, a processor in the central server being arranged to manage a plurality of modules in conjunction with the instructions stored on the memory device, the modules comprising:
a. An attributes & Levels Determination module configured to
determine a historical set of attributes and levels for scenario
analysis based on stored historical data associated with one or
more existing products in said central repository;
b. an input module configured to accept a set of new attributes from
a user and its associated levels therefor; thereby generating
updated set of attributes and levels;

c. A design of experiment module adapted to design a profile for
said product and a choice model thereof, wherein said model
enables selection of a sub-set of updated attributes and levels in
context with said designed profile;
d. a Hierarchical Bayesian analysis module configured to perform a
Hierarchical Bayesian model analysis on the selected sub-set of
updated attributes and levels to determine the part worth utilities
of each selected sub-set of attribute at different levels; and
e. A simulator configured to simulate plurality of scenarios by
calibrating the selected attributes at different selected levels and
considering part-worth utilities thereof to determine the best
possible scenario depicting the performance of the product.
7. The system of claim 6, wherein said choice card is designed to determine the preferences of multiple users related to existing products analyzed at said attributes and levels.
8. The system of claim 6, wherein said design of profile and choice card is enabled by means of fractional factorial design that enables to reduce the number of scenarios and correctly estimate the part worth utility of each product.
9. The system of claim 6, wherein the new attributes and levels are specified by the user and hierarchical Bayesian model is used in order to predict the part worth utility of each new attribute.