ITECH7407 – DATA ANALYTICS
Submitted By -
TABLE OF CONTENTS:
1. PURPOSE
2. BACKGROUND & INFORMATION
3. DATA MINING
4. RESEARCH
5. RECOMMENDATION FOR CEO
6. COVER LETTER
7. REFERENCES
1. PURPOSE
The purpose of undertaking this activity is to understand the Business Intelligence Techniques by considering a particular case study.
As we all are on the verge of Industrialization 5.0, where the Artificial Intelligence will take over the contemporary way of working. Business Intelligence & Predictive analysis would be the key procedures in Standard Operating Procedure which will further enhance the decision making power for management. (Yanqing, 2020)
Business Intelligence is the functionality to capture the transactional data of any procedure within the organizations, e.g. Sales, Supply Chain, Logistics, Procurement etc., understand the pattern & provides the forecast through predictive analysis.
This forecast analysis further helps the organization & its management to analyse the gaps within the procedures, plug in those gaps for improving the overall efficiency of the organization.
2. BACKGROUND & INFORMATION
We have considered the data set of Sales transaction of a Superstore. The data is segregated into level of order priority as Critical, High, Medium & Low.
It is assumed that higher the level of criticality, higher would be the premium charges on that particular product in terms of higher shipping cost, less discount, etc.
Based on the criticality of the order, each transaction is having a unique Order ID. This order ID further contains the unique Product ID along with the unique item description corresponding to that order ID. It also contains the Order date & shipping date & mode of shipment, i.e. the date when material is despatched. The overall sales data is marked with a specific categorization & sub categorization of the product. These categorization & sub-categorizations basically define the type of expenses like Technological product, Furniture Supplies, office supplies like stationary items etc.
Further, it contains the customer details like, customer name, its segment, whether it a Corporate customer, Household Customer, Consumer. Etc., its State, Country, It will help us to understand the consumer behaviour across various regions.
In terms of commercials, it contains the sale price, quantity ordered, discounts offered based on customer & quantity pattern, shipping cost etc.
We will analyse the data set through SAP Predictive Analysis & will draw some meaningful insights which will further enhance the business situation

Figure 1: Sample image of dataset Superstore Sales
3. DATA MINING
The whole data of sales transaction is uploaded in SAP Predictive Analysis tool. SAP predictive analysis is one of the powerful BI tool & also known as flag bearer of SAP family. The key functionalities of the tool are Business Intelligence, Planning, Reporting, Forecast analysis, visualizations & many more. (Help.sap.com)
Figure 2: Data Mining in Predictive Analysis
Data Modelling is done once the data is cleaned into a specific structure. The data has been categorized into measures & dimensions based on the nature of data, For example, order date & shipment date are converted into a date format, sale price, discount, shipping charges converted into numeric, description & order ID into text format.
Further, this data is converted into clusters & simpler forms. We will understand the with the help of various insights which we could draw with the help of sale transaction date set.
4. RESEARCH
SCOPE 1 – Classification of Dataset
In this section of our analysis we will be exploring the Percentage share of each sub-category that is being contributed to the total sales percentage of the corresponding categories. This will help us to know which categories are the most which are contributing the most to the total sales for any particular category. For this particular analysis we will be using the horizontal bar chart where the x-axis we have sales percentages and on the Y-axis, we have category and the various subcategories under that particular category.
Figure 3: Classification of Dataset
Conclusion (Category wise) –
It can be seen from the above analysis that -
1) Furniture - Chairs is having the maximum share in the total sales percentage, but it can be seen that there is not much difference between the Chairs and the Book cases sales share.
2) Office Supplies - Storage is having the maximum share in the total sales under office supplies category. Additionally, around 50% of the share in Office supplies is accounted by Storage and Appliances. As can be seen from the above horizontal bar chart that there is a huge gap in the contribution between the first Two and the others subcategories sales share under Office category. These needs to be investigated and find out the reason for such a huge gap.
3) Technology - Phones is having the maximum share in the sales share under Technology category and followed by the Copiers- which together account for 60% of total Technology sales. As can be seen from the above that around 60% of the sales of technology category is accounted by the Copiers and Phones. The percentage share of the remaining categories under technology is negligible or very less and needs to be focussed so as to increase their share as well.
Recommendation –
From the above analysis it can be concluded that there is a very high difference in the share that is being contributed by different subcategories in their corresponding categories. This huge gap among the various subcategories needs to be further drill down so as to be able to find the real reason for this particular huge...