FORM NO 2
THE PATENTS ACT, 1970
(39 OF 1970)
&
The Patents Rules, 2003
COMPLETE SPECIFICATION
[See sections 10 and Rule 13]

1. TITLE SYSTEM AND TECHNOLOGY DESIGN FOR COMPUTATIONAL AND PROCESSING OF DATA AND VISUALIZATION

2. APPLICANT(S) PLANET NEXTGEN TECHNOLOGIES INDIA PRIVATE LIMITED
5, Jyoti Wire House, 23 A – Shah Industrial Estate, Off Veera Desai Road, Andheri (W), Mumbai - 400053 India.
3. PREAMBLE TO THE DESCRITION
COMPLETE
The following specification particularly describes the invention and the manner in which it is to be performed

Title
SYSTEM AND TECHNOLOGY DESIGN FOR COMPUTATIONAL AND PROCESSING OF DATA AND VISUALIZATION
FIELDS OF INVENTION
The present invention is in the field of System and Technology design for computational, processing of data and forecasting data for business. The product and process are for the use of business individual and Organisations, for data forecasting, event management and Promotion planning to analyse, predict and visualize information and includes data warehouse, data mining, Data Harmonisation, data visualisation, business analytics or business intelligence.
BACKGROUND DISCUSSIONS
With the new era of professional digital Business System, the business approach is focus on to provide an sustainable forecasting tool to regulate demand and supply and thereby achieve efficient Inventory management, promotion planning, event management and thereby focued on driving actionable and process automation, that would target to enhance customer sales and improve investing management in the most economic manner
The present invention is attempt to provide for a system with a tool for forecasting demand data for business organization and thereby creating an supply chain management and in inventory management system.
OBJECT OF THE INVENTION
The Primary Object of the Invention is to provide for a tool and system for forecasting on daily, weekly and on monthly basis for demand and supply of the products
It is yet another of object of the invention to provide a system based on machine learning system that helps to manage inventory based on events, periods like seasonality and themby achieve resource allocation with less cost and increase revenue thereby provide an economic significance
The objective of the invention is to provide a system and accompanying methods for accurately forecasting future demand for many products and product types in many markets. Also, another objective of the tool to provide a system and related methods that enables organizations to produce and compare alternative models of forecasted demand in order to constantly improve demand-forecasting capabilities.
It is also the object of the invention to provide for tool to provide a system and accompanying methods producing and identifying optimal demand forecasts that take into account independent causal factors, such as new competitive products that will impact upon future demand.Further, the tool to a system and related method whereby users can easily account for current demand trends without having to produce an entirely new forecast.

It is also object of the invention on to provide a system and related methods that allows users to produce and compare forecasts more easily for related products, products within related markets and for products, taking into account different forecasting models. The tool helps users to determine the forecast algorithm that best suits a given problem/industry/business from the available history streams of demand data by enabling the production of various forecasts for comparison with one another and, eventually, with incoming data relating to actual demand.
By employing multiple models for a single product or location, the tool enables multiple-scenario comparisons and analysis by letting users create forecasts from multiple history streams (for example, shipments data, point-of-sale data, customer order data, return data, etc.) with various alternative forecast algorithm.
Further, the tool uses an automatic tuning feature to assist users in determining optimal parameter settings for a given forecasting algorithm to produce the best possible forecasting model. This feature significantly reduces the time and resources necessary for users to produce an acceptable demand forecast from history streams of demand data.
Also, the tool provides a system and method whereby appropriate demand responses can be dynamically forecasted whenever given events occur, such as when a competitor lowers the price on a particular product (such as foran Event or festival), or when the user's company is launching new sales and marketing campaigns. Utilizing the multiple model framework according to the tool allows the selection, customization, and fine tuning of the appropriate forecasting algorithm to apply to a given demand problem by producing a plurality of forecasts, each forecast taking into account different demand causal factors and demand data history and identifying and optimal forecast for adoption. According to such condition, users can accurately predict customers' buying plans and thereby forecast the demand for a product or product line by identifying and quantifying key variables that drive demand from the available history streams of demand data.
The demand forecasting capabilities of the tool can be distributed across an entire enterprise or alternatively can be used regionally for specific geographic areas of an enterprise using demographical data or weather data.The tool also provides intelligent event modelling capabilities and allows for management overrides and forecast adjustments, given newly received data regarding changes in actual demand. As such, event data is incorporated into forecast models on an ongoing basis to help adjust forecast decisions on the fly and refine the forecasting process for future endeavours. By including both market planning and demand planning capabilities, it links product mix, promotion, and price analyses with traditional demand forecasting.
One of the ordinary skills in the art will appreciate that the demand forecasting algorithms may be modified to take into account specific needs and/or problems encountered in particular industries or situations. Thus, the illustrative algorithms should not be construed to limit the tool as is claimed.
Although the tool is preferably implemented in software, this is not a limitation as it can be implemented in hardware or in various combinations of hardware and software, without departing from the scope of the invention. Modifications and substitutions of ordinary skill in the art are considered to be within the scope of the tool, which is not to be limited except by the claims that follow.

DETAIL DESCRIPTION OF INVENTION
The present invention is the tool seeks to Demand Plan, Cost manage, Inventory manage and Optimise Inventory by forecasting its demand over the time and by using the same methods for simulating the distribution of the same over a fixed period time.
The tool includes an inventory management system that utilizes the machine learning under a continuous review model.
The continuous review model determines two quantities for each item, a reorder point and an order quantity. When the inventory is about to exhaust and reach a reorder point, an order is placed for an amount equal to the one which was placed formerly to replenish stock.
This tool helps in not only forecasting order placed for its next renewal but also calculates a forecast pattern for entire distribution cycle by using a periodic review model, which determines a review interval and “Order-Up-To Level.”
The status of the on-hand inventory is checked only at fixed intervals called “review periods,” i.e. every 1 month. There will be shrinkage in on-hand inventory during a review period. when this occurs, the contingency given to user would to go with an option of placing a replenishment order large enough to return the total inventory.
In addition to providing the above re-order information for continuous and periodic review models, the tool also provides performance measures. Performance measures depend not only on the demand forecasts, but also on economic, operational and service level factors or values supplied. The economic factors include costs for holding and ordering inventory. The operational factors specify what happens when demand exceeds on-hand inventory, i.e., lost orders or back orders. The service level factors include a choice of service level criterion and the minimum acceptable value for that criterion.
The approach taken for forecasting intermittent demand in accordance with this tool involves the use of various machine learning algorithms. The machine learning algorithms can be defined as a method of computational inference based on creating simulated replicates by re-sampling the historical data.The machine learning algorithms creates a plurality of sets of predicted future values, wherein each set of predicted future values is based on the sampling of historical data. The plurality of sets (or their sums) creates a distribution that can thereafter be statistically analysed or inputted into an inventory control system. The advantage of utilizing machine learning algorithms for forecasting intermittent data is that it not only reproduces standard results in standard situations, but it also applies to situations that lack analytical formulas or that violate the assumptions on which available formulas are based.
The tool relates to a system and method for demand forecasting, as discussed in detail below in connection with FIGS. 1-4.
The basic assumptions of a time series model are that there is information about the past, this information can be quantified in the form of data, and the pattern of the past will continue into the future. The tool is adjustable depending on the needs of the user. The time series model of a product's daily volume depends on many factors, which can be grouped into execution factors,

trend/festive/seasonality, and market factors. Execution factors (e.g. market, weekdays, number of stores, where the product is on sale, when and where products are sold out, etc.) can be adjusted macroscopically in the following year, but are hard to quantify and predict. Factors are mainly determined by the inventory of the product and can be quantified. The strategy can also be planned and adjusted in real time. Market factors (e.g., event, trend, climate, time on sale, holidays, etc.) reflect the acceptance level of the consumers to the products, the change of the purchasing power, and the market saturation level. Market factors vary with time and product so that different products might behave differently under the same circumstances.
The FIG. 2 is a flowchart showing processing steps carried out by the engine 10 for developing a time series model. Starting in step, a quantity selling rate is defined to represent the demand in order to eliminate the influences from execution factors on sales volume. The quantity rate of certain product within a given period is defined as the number of products sold per store per day with stock.
FIG. 7 showshypothetical previous, current, and estimated volumes. Such data could be used for inventory optimization in the executing and adjusting stages. Once the modelling parameters are available, the actual volume can be reproduced by applying the execution factors to the quantity, i.e., multiplying the quantity by the number of stores across the markets and the number of flow days. The forecasting can then be used to optimize the purchase plan and improve the inventory management in advance, and optimize the pricing strategy during operation. Thus, the forecasting results can be used as a guide for price adjustments.
The Machine Learning (ML) provides a mathematical framework for decision-making. The outcome has a random is a component that is under the control of the decision maker.
The throughput of this Machine Learning Process is to pre-determine the market behaviour to certain conditions
This ML approach will help in Demand Planning, Cost management, Inventory management, forecasting and Inventoryoptimisation, Most Importantly it would also be focused on driving process automation that is targeted to increase the organic growth of the company.
The system and product can provide personalized customer relationship by monitoring, managing, evaluating highly complex processes.
The ML provides a mathematical framework for decision-making problems where the outcome has a random component and a component that is under the control of the decision maker. FIG. 7-11 shows plot illustrating processing steps carried out by the system for conducting ananalysis to calculate a demand forecast for a product. In many industries (e.g., consumption pattern and retail industry), many products are ‘like’or ‘dis-like’goods without historical sales records, which can be described by their physical properties and other derived properties. For example, for freezer-temperature(e.g., meat, chicken patty, etc.), the dryer -temperature type (e.g., vegetable). The assumption of analysis is that if two products share the same key characteristics, their model parameters should be similar, if not the same (i.e., more characteristics the products share, likelier for their parameters to be the same). To get an initial estimate of the model parameters to forecast the demand for a brand new product, analysis is performed on the intrinsic properties of the products.

2. Detailed Description
The product is a business analytics tool that forecast food productspresent within a store on a monthly, weekly and daily basis. The product uses various machine learning algorithms that monitor, analyse, optimise and give results at 95% confidence level.
Time Series Forecasting provides an interactive user interface to visualise and forecast time series. The user interface allows users to compare fitted time series models and forecasts with several algorithms including.The data is fetch from the system by setting query in Oracle/SQL data base. To carry out statistical computing that needs very advanced and complex SQL queries that pull data from Database platform (NetSuite). This database is a Relational database system that are stored in a normalized format. R connects to relational databases like MySql, Oracle, Sql server and fetch records as a data frame. After fetching the data from server to R, the statistical computing is done and Forecast models are develop using Machine learning approach. The models provide an interactive user interface to visualise and forecast time series. The tool allows users to compare fitted time series models and forecasts with several algorithms including:
• Holt-Winters (HW): The Holt-Winters exponential smoothing is applied for data exhibits for trends and seasonality. Holt-Winters exponential smoothing estimates the level, slope and seasonal component at the current time point. This algorithm is applied for forecasting sales of product during festivals and events to allow user to adjust for surge in inventory.
• ARIMA with Fourier Transform (FTARIMA): Autoregressive Integrated Moving Average (ARIMA) models applied is an explicit statistical model for the irregular component of a time series that allows for non-zero autocorrelations in the irregular component. This is method is applied for estimating the impact and forecasting for seasonality such as month end and week end. ARIMA models are defined for stationary time series. Therefore, if start off with a non-stationary time series, it will first need to 'difference' the time series until the obtain a stationary time series. A Fourier series approach where the seasonal pattern is modelled using
The advantages of this approach are:
• It allows any length of seasonality.
• For data with more than one seasonal period, you can include Fourier terms of different frequencies;
• The seasonal pattern is smooth for small values.
• The short-term dynamics are easily handled with a simple ARMA error.
• TBATS: TBATS is an exponential smoothing model with Box-Cox transformation, ARMA, trend
and seasonal components. It tunes its parameters automatically. Though a very good model
that operate in a completely automated manner. It means the application has no manual
intervention and perform for better forecasting.

• HYBRID: The hybrid Model function fits multiple individual model specifications to allow easy creation of ensemble forecasts. It provides the best result for the product as since it works on the data that is already delivering the accuracy of 85% and help in optimising that accuracy by 95% across all product on monthly weekly and Daily basis.
3. Measures of Forecast Accuracy (Performance Metrics)
In order the tool described above, over 3 million data series of actual intermittent data were studied. The data is summarized in the tablel. Assessing the performance of forecasting methods for intermittent demand required the development of new measures. When assessing forecasts of smooth demand, it is conventional to use such accuracy measures as the mean squared error (MSE) or the mean absolute percentage error (MAPE). The latter is generally preferred when one wants to assess the performance of alternative forecasting methods over many items, because its relative, scale-free measure can be averaged naturally across series with different mean levels.
SUMMARY OF THE INVENTION
The objective of the invention is to provide a system and accompanying methods for accurately forecasting future demand for many products and product types in many markets. Also, another objective of the tool to provide a system and related methods that enables organizations to produce and compare alternative models of forecasted demand in order to constantly improve demand-forecasting capabilities.
It is also the object of the invention to provide for tool to provide a system and accompanying methods producing and identifying optimal demand forecasts that take into account independent causal factors, such as new competitive products that will impact upon future demand.Further, the tool to a system and related method whereby users can easily account for current demand trends without having to produce an entirely new forecast.
It is also object of the invention on to provide a system and related methods that allows users to produce and compare forecasts more easily for related products, products within related markets and for products, taking into account different forecasting models. The tool helps users to determine the forecast algorithm that best suits a given problem/industry/business from the available history streams of demand data by enabling the production of various forecasts for comparison with one another and, eventually, with incoming data relating to actual demand.
By employing multiple models for a single product or location, the tool enables multiple-scenario comparisons and analysis by letting users create forecasts from multiple history streams (for example, shipments data, point-of-sale data, customer order data, return data, etc.) with various alternative forecast algorithm.
Further, the tool uses an automatic tuning feature to assist users in determining optimal parameter settings for a given forecasting algorithm to produce the best possible forecasting model. This feature significantly reduces the time and resources necessary for users to produce an acceptable demand forecast from history streams of demand data.
Also, the tool provides a system and method whereby appropriate demand responses can be dynamically forecasted whenever given events occur, such as when a competitor lowers the price on

a particular product (such as foran Event or festival), or when the user's company is launching new sales and marketing campaigns. Utilizing the multiple model framework according to the tool allows the selection, customization, and fine tuning of the appropriate forecasting algorithm to apply to a given demand problem by producing a plurality of forecasts, each forecast taking into account different demand causal factors and demand data history and identifying and optimal forecast for adoption. According to such condition, users can accurately predict customers' buying plans and thereby forecast the demand for a product or product line by identifying and quantifying key variables that drive demand from the available history streams of demand data.
The demand forecasting capabilities of the tool can be distributed across an entire enterprise or alternatively can be used regionally for specific geographic areas of an enterprise using demographical data or weather data. The tool also provides intelligent event modelling capabilities and allows for management overrides and forecast adjustments, given newly received data regarding changes in actual demand. As such, event data is incorporated into forecast models on an ongoing basis to help adjust forecast decisions on the fly and refine the forecasting process for future endeavours. By including both market planning and demand planning capabilities, it links product mix, promotion, and price analyses with traditional demand forecasting.
One of the ordinary skills in the art will appreciate that the demand forecasting algorithms may be modified to take into account specific needs and/or problems encountered in particular industries or situations. Thus, the illustrative algorithms should not be construed to limit the tool as is claimed.
Although the tool is preferably implemented in software, this is not a limitation as it can be implemented in hardware or in various combinations of hardware and software, without departing from the scope of the invention. Modifications and substitutions of ordinary skill in the art are considered to be within the scope of the tool, which is not to be limited except by the claims that follow.

BRIEF DESCRIPTION OF THE DRAWINGS and TABLES
The accompanying drawings, which are included to provide further understanding of the tool and are incorporated in and constitute a part of this specification, illustrate the tool and together with the description serve to explain the principles of the tool. In the drawings with reference numbers representing corresponding parts throughout:
FIGURE. 1 is a schematic diagram and isthe architecture of the tool that explain the overall process steps performed to produce a demand forecast including preparing for and performing a demand planning cycle andhow demand forecasts can be organized by the tool and the alternative forecasting algorithms and various history streams of demand data can be combined according to the tool create a multiple model framework. It also shows the operational aspects and interactions of an electronic demand forecasting system according to tool.
FIGURE. 2 is a flow diagram depicting how the logic whereby the demand forecasting system of the tool creates a forecast for a given demand forecasting unit from history stream and user inputted level, trend, seasonal effects,causal factors according to the tool.
FIGURE. 3 is a combination schematic and flow diagram depicting the manner by which the various coefficients for a forecasting algorithm are generated according to preferred tool.
FIGURE. 4 is a plot diagram depicting the demand data sample with respect to time according to the tool.
FIGURE. 5 is a diagram depicting the demand data sample in to the time series sampleaccording to the tool.
FIGURE. 6 is a plot diagram depicting the data sample decompose in to time series demand data and segregate in to observe data, trend data, seasonality data and random data as according to tool.
FIGURE. 7 is a plot diagram depicting the conversion of data sample into time series demand data sample according to the tool.
FIGURE. Sis a plot diagram depicting adjustments in the time series demand data sample according to the tool.
FIGURE. 9 is a plot diagram depicting fitted data in to the time series demand data sampleaccording to the tool.
FIGURE. 10 is a plot diagram depicting the forecast trend for the time series demand data sample according to the tool.
FIGURE. 11 is a plot diagram depictingthe range along with the forecast trend for the time series
demand data sampleaccording to the tool.

Table.1 is a table depicting the performance meter for the forecast trend for the time series demand
data sample according to the tool.

We Claims:
1. A system for providing forecasting tool for demand and supply of product using machine
learning system comprising of:
System based machine learning tool to forecast preferably on daily, weekly or on monthly basis for demand and supply of the products, also forecasting inventory based on events, periodicals, enabling resource allocation and control cost also providing a comparative alternative model for forecasting demand and thereby improving forecasting capabilities, also providing system of identifying optimal demand forecasting system to analyze comparative models with related markets, that enables to forecast in multiple scenario and analysis.
2. A system as claimed in claim 1 in , wherein the operations comprise:
a tool to regulate demand and supply and achieve an efficient inventory management, promotion planning, event management.
3. A system as claimed in claim 1 and 2 above, wherein the operations comprise: a system that allow users to produce and compare forecast more easily for related products, products within related markets and for products taking into account different forecasting models.
4. A system as claimed in claim 1 to 3, wherein the operations comprise:
A system and accompanying methods for accurately forecasting future demand for many products and product types in many markets.
5. A system as claimed in claim 1 to 4, wherein the operations comprise: A tool that enables multiple-scenario comparisons and analysis by letting users create forecasts from multiple history streams with various alternative forecast algorithm. /
6. A computer -implemented system for access management using machine learning providing a mathematical frame work for decision making in all respects.