SMART PREDICTION SYSTEM AND METHODS FOR RIGHT PRODUCT SELECTION

Abstract

This invention focuses on all the manufacturing industries who are involved in taking decision regarding right product selection and it discloses an intelligent big data system and method for such selection. Various manufacturing Industries generate various kinds of data which is bulky and in structured unstructured and 10 semi-structured form. This data collected and analyzed with the help of big data analytical technique which will provide an insight to the industries for right product selection.

Specification

Description:FIELD OF THE INVENTION
The invention related with the creation 5 of model using big data analytical
techniques. The big data will be processed and with the help of various data
analytical tools a model can be built along with some visualization to gain an insight
for right product selection.
10 BACKGROUND OF THE INVENTION
In the modern industrial landscape, product manufacturing is a highly complex and
costly process, involving significant investments in terms of time, capital, and
resources. The challenge for businesses is to develop and launch the right products
that meet market demands while optimizing costs and maximizing profitability.
15 One of the key difficulties faced by industries, particularly in sectors like
pharmaceuticals, consumer goods, and technology, is the identification and
selection of the most suitable product for development and eventual
commercialization.
20 Manufacturing industries today operate in a data-intensive environment, where vast
amounts of information are generated daily. This data originates from various
sources, including production processes, supply chains, customer feedback, sales
records, market research, and competitor analysis. The volume, velocity, and
variety of this data are increasing exponentially due to advances in digital
25 technologies, making it difficult for businesses to manually process, analyze, and
draw meaningful insights from this information.
Despite the availability of a wealth of data, many industries still struggle to make
informed and timely decisions regarding product development. This is because
30 traditional decision-making methods often rely on subjective judgment or outdated
analytical tools that are inadequate for processing the large, complex datasets being
3
generated today. As a result, businesses may develop products that are either poorly
timed or misaligned with market needs, leading to reduced profitability and
increased risk of failure.
The advent of big data analytics has revolutionized 5 the way industries can approach
product development. Big data analytics refers to the use of advanced data
processing techniques and tools to analyze large datasets and uncover hidden
patterns, correlations, and insights that would otherwise be difficult to detect. By
leveraging big data, companies can gain a deeper understanding of market trends,
10 customer preferences, and competitive dynamics, enabling them to make datadriven
decisions about which products to develop and when to bring them to market.
Predictive analytics, in particular, has emerged as a powerful tool for guiding
product selection. Predictive analytics uses historical data, machine learning
15 algorithms, and statistical models to forecast future outcomes, such as product
demand, sales trends, and pricing fluctuations. With the help of predictive analytics,
industries can make more accurate predictions about the potential success of a
product, as well as identify the most profitable markets and customer segments to
target.
20
While various analytical tools and techniques are available, there remains a need
for a comprehensive system that integrates multiple forms of data analysis-such
as descriptive, diagnostic, predictive, and prescriptive analytics-to assist
industries in making faster, more accurate, and more informed decisions about
25 product selection. Such a system would allow businesses to not only predict future
product performance but also optimize the entire product development lifecycle,
from concept to commercialization.
In light of these challenges, the present invention addresses the need for an
30 intelligent, data-driven model that uses big data analytics and predictive modeling
to assist businesses in selecting the right products for development. By combining
4
various big data analytical techniques, including forecasting, classification,
clustering, and time-series analysis, this model provides a comprehensive solution
for analyzing product-related data and making informed product development
decisions. The invention aims to significantly reduce the risk associated with
product selection and ensure that industries 5 can bring the right products to market
at the right time, thus optimizing both profitability and market success.
SUMMARY OF THE INVENTION
The primary aim of this invention is to develop an intelligent predictive model using
big data analytical techniques. This model will serve as a guide for the right product
10 selection, fulfilling the following objectives:
1. To capture data from various sectors and identify the suitable sector for the
product.
2. To analyze data using big data analytical techniques.
3. To provide insights to companies regarding the right product selection.
15 4. To identify the target market for profitability.
5. To study sales trends of products.
6. To predict the product timeline.
7. To forecast future product pricing.
8. To identify the number of competitors in the market selling the same
20 product.
The invention employs several big data analytical techniques such as descriptive
analytics, diagnostic analytics, predictive analytics, and prescriptive analytics.
Different predictive models, including forecasting, classification, clustering, outlier
detection, and time series models, are used to achieve these objectives.
25 In order to achieve above objectives the technical solution adopted by the present
invention is as follows.
Data Collection-Data can be collected through various mediums like observations,
personal interviews, and online mode or through various books which will contain
5
recent information about the drug. Quantitative data can be collected through above
mentioned methods and different data cleaning methods applied to clean the data.
Data Cleaning-A method which is responsible for removing bad, noisy,
inconsistent, duplicate, missing or incomplete data in a dataset is called data
cleaning. By making use of some predefined 5 software data cleaning can be
achieved.
Data Processing-The data which is collected and cleaned can be considered as a
data for processing. The data collected and cleaned will be represented in some file.
Furthermore this data will be used to process raw data which will gets converted
10 into a meaningful form through an electronic medium. The data stored in a file will
be processed by the application and analytical software's
Data Visualization- The processed data can be represented and can be visualized
by means of data analytical software.
In order to achieve forth to eighth objective different big data analytical techniques
15 are used. There are mainly 4 types of big data analytical techniques namely
descriptive analytics, diagnostic analysis, predictive analytics and prescriptive
analytics etc. Combinations of these techniques will be used to achieve the
objectives.
More specifically various types of predictive models like forecasting, classification,
20 clustering, outlier, time series model will be used to achieve forth to eighth
objectives.
BRIEF DESCRIPTION OF THE DRAWINGS
Fig 1. Big data in pharmaceutical
Fig 2. Traditional drug selection process.
25 Fig 3. Data analysis
6
DETAILED DESCRIPTION OF THE INVENTION
The following is a detailed description of embodiments of the present disclosure.
The embodiments are in such detail as to clearly communicate the disclosure.
However, the amount of detail offered is not intended to limit the anticipated
variations of embodiments; on the 5 contrary, the intention is to cover all
modifications, equivalents, and alternatives falling within the spirit and scope of the
present disclosure as defined by the appended claims.
Unless the context requires otherwise, throughout the specification which follow,
the word "comprise" and variations thereof, such as, "comprises" and "comprising"
10 are to be construed in an open, inclusive sense that is as "including, but not limited
to."
Reference throughout this specification to "one embodiment" or "an embodiment"
means that a particular feature, structure or characteristic described in connection
with the embodiment is included in at least one embodiment. Thus, the appearances
15 of the phrases "in one embodiment" or "in an embodiment" in various places
throughout this specification are not necessarily all referring to the same
embodiment. Furthermore, the particular features, structures, or characteristics may
be combined in any suitable manner in one or more embodiments.
The headings and abstract of the invention provided herein are for convenience only
20 and do not interpret the scope or meaning of the embodiments.
The invention related with the creation of model using big data analytical
techniques. The big data will be processed and with the help of various data
analytical tools a model can be built along with some visualization to gain an insight
for right product selection.
25
Predictive Analytics - The term predictive analytics refers to the use of statistics
and modeling techniques to make predictions about future outcomes and
performance. This type of analysis considers both the historical and present data to
7
make predictions of the future. It uses the concepts of Artificial Intelligence (AI),
data mining and machine learning algorithms to analyze current data and make
predictions about the future. Some of the examples of predictive analysis include
rain forecasting, understanding customer likes and dislikes, make predictions about
market etc. In this paper we will be 5 using this predictive analysis to make a right
product selection. Various types of predictive models will be used while performing
data analytics those are as follows
Forecast Model - In this model we estimate the values of new data from historical
data available. It accepts multiple parameters as input and predicts the output. e.g.
10 It takes into consideration historical sales data and predicts the future sale based on
the inputs.
Classification Model - These models work by categorizing information based on
historical data. Classification models are best to answer yes or no questions,
providing broad analysis that's helpful for guiding decisive action. These models
15 can answer questions such as the statements where from 2 options we have to select
one option.
Clustering Model - This model sorts the data into groups based on common
attributes. It works by grouping things with shared characteristics or behaviors and
plans strategies for each group at a larger scale
20 Outliers Model - The outliers' model works with anomalous data entries within a
dataset. It can identify anomalous figures either by themselves or in conjunction
with other numbers and categories.
Time Series Model - It accepts time as an input parameter to develop numeric metric
that will predict trends within a specified period. This model can be useful to
25 identify the timeline of the product.
Python for Data Analysis - Python is interpreted, high level, object oriented
programming language which has its rigorous existence in the field of data science
8
and analytics. It provides a set of robust and standard libraries which are used for
analyzing and visualizing the data. It builds better analytics tools which help in
developing machine learning models, web services, data mining, classification etc.
Python Libraries for Data Analytics
1. NumPy - It supports n-dimensional 5 array and provides numerical computing
tools. It's useful for linear algebra
2. Pandas - It provides functions to handle missing data and performs
mathematical operations. Provides tools for data manipulation functions like
reshaping, merging, sorting, slicing, aggregation etc.
10 3. Matplotlib - It is a cross platform, data visualization and graphical plotting
library used for plotting data points and creating interactive visualization of the
data.
4. Scipy - This library is used for scientific computing. It is a collection of
algorithms for linear algebra, differential equations, numerical integration,
15 optimization, statistics and more. It contains modules for optimization, linear
algebra, integration, interpolation, special functions, signal and image
processing.
5. SciKit-Learn - Provides machine learning algorithms such as classification,
regression, clustering, model validation etc. ([5] Ritu Ratra, Dec 2019)
20 Power BI - Microsoft's Power BI is an open source; cloud based analytical service
used for visualizing, exploring and extracting insights from data. It's an approach
used for transforming raw data into meaningful business insights. It also has a
feature called Quick Insights which is built on a growing set of advanced analytical
algorithms. The analysis of an uploaded dataset can be achieved within a single
25 click. Accepting even Excel spreadsheets as input, Power BI is easy to use and ripe
for adoption as a platform for Research Data Analysis. The analytical reports and
visuals generated by this tool can be shared easily.
9
Thus python can be used as a data analysis for quicker decision making and finding
better opportunities and with power BI visualizations are possible so that companies
can take help of these visualizations which will provide them a guideline for right
product selection
A manufacturing Industry is a commercial 5 business whose focus is to research,
develop, market and distribute various kinds of products. The products possess
various properties and can gives rise to huge dataset. In order to manage such huge
volume or bulky data some big data analytics techniques can be used. Effective
analysis of this data will aid the manufacturing companies to make the decision
10 regarding right product selection.
e. g. Every pharmaceutical Industries criteria for drug selection is different but
roughly many
Industries follow the below mentioned process for drug selection.
Marketing Team- The team of Medical Representatives are acting as a key point of
15 contact between pharmaceutical companies and health care professional. Their role
is to identify a demand for an existing product. In order to identify new business
opportunity of drugs into the market they interact with different types of people like
doctors, nurses, pharmacy owners, hospitals etc. Based on the inputs taken from
these people a report is presented to business development team for further process.
20 Marketing team gives input regarding the drug to business development team based
on the below parameters
1. Brand Name- It refers to the name of the pharmaceutical company currently
involved in drug production.
2. Volume-It refers to the volume of production of drugs
25 3. Value-The current market value of the drug
4. Suppliers-These are the distributors involved in supplying of the drug.
Business Development Team- The major role of business development team is to
maintain robust products pipeline and ensure the launch of the right product at right
10
time. The business development team takes into consideration the input generated
from marketing team and carry out analysis based on the following points
1. Historical data- It considers the historical data related to that drug. It also takes
help of social media in order to gain information regarding the sales trend of
5 that drug.
2. Volume and value- It performs the analysis to determine the volume and value
of the drug under consideration.
3. Selling Prize-Selling prize of the drug under consideration is decided by
performing its comparison with the drug in the market. This prize must be lower
10 than the prize of the drug in the market.
4. Cost of goods sold (Cogs)-It identifies the actual cost of manufacturing of the
drug and it coveys this cost to Research and Development department.
Finance Team- Based on the analysis done by business development team the
15 finance team identifies approximate Investments, Return on Investments (ROI), and
payback period of the drug under consideration.
Senior Management - The above analysis done by marketing team, business
development team and finance team is put forth in front of senior management team
20 of the company including owner, experts, board of directors etc. After a review
senior management recommends the DP and API to be developed to the Research
and Development team.
Research & Development- Research and Development department of any
25 pharmaceutical Industry plays a vital role for attainment of company's growth.
Introducing a new drug into the market is expensive and risky process therefore
proper analysis of the emerging drug selection is very necessary. Any decision
taken by this department has their long term consequences and its impact is directly
proportional into the market. So based on the inputs received from business
30 development and finance team the Research and Development team carries out
following feasibility studies
11
1. Cost of goods sold estimates(Cogs)-It calculates that whether it is feasible to
develop the emerging product or drug in the same cost or lesser than the cost
analyzed by business development team.
2. Return on Investments (ROI)-The drug development is expensive and time
consuming process. It takes near about 5 2 to 2.5 years to launch that drug into
the market. So before investments we should also think about the Returns on
the investments.
Apart from these parameters it also performs analysis regarding number of players,
10 sales trend etc. After successful analysis the drug is filled for the patent approval
along with a stability check and then approved drug is taken into consideration for
development. , Claims:1. An intelligent predictive analytical big data system comprising:
2. The intelligent predictive analytical big data system as claimed in claim 1,
wherein system comprising
5 a. a processor;
b. at least one data analytical software;
c. at least one visualization software; wherein the system is configured to
perform data collection, cleaning, processing, and visualization, thereby
enabling the analysis and prediction of data trends.
10 3. The intelligent predictive analytical big data system as claimed in claim 1,
wherein the model to be developed comprises:
a. data collection from multiple sources;
b. data filtration for noise removal;
c. data analysis to extract meaningful insights;
15 d. and inference generation for predictive analysis.
4. The intelligent predictive analytical big data system as claimed in claim 1,
wherein a dashboard is created using the data analytical and visualization
tool, wherein the dashboard is configured with a product dataset and
provides insights that aid in decision-making for selecting the right product.

Documents