Professional Documents
Culture Documents
Final Paper Spring 2021 - BusinessIntelligence by Shariq Ahmed Khan 58549
Final Paper Spring 2021 - BusinessIntelligence by Shariq Ahmed Khan 58549
Final Paper Spring 2021 - BusinessIntelligence by Shariq Ahmed Khan 58549
Instructions:
• Filling out Student-ID and Student-Name on exam header is mandatory.
• Do not remove or change any part of exam header or question paper.
• Write down your answers in given space or at the end of exam paper with proper title “Answer for
Question# _ _”.
• Answers should be formatted correctly (font size, alignment and etc.)
• Handwritten text or image should be on A4 size page with clear visibility of contents.
• Only PDF format is accepted (Student are advise to install necessary software)
• In case of CHEATING, COPIED material or any unfair means would result in negative marking or
ZERO.
A mandatory recorded viva session will be conducted to ascertain the quality of answer scripts where
deemed necessary.
Caution: NO EXTRA TIME. If you failed to upload answer sheet on LMS (in PDF format) within 03 hours
limit, you would be considered as ABSENT/FAILED.
Page 1 of 14
Question 1: Agha Super Mart Chain 10 marks
Agha super mart is a large grocery chain. Their business has 100 grocery stores spread over a four provinces.
Each of the stores has a full complement of departments, including grocery, frozen foods, dairy, meat, produce,
bakery, floral, and health/beauty aids. Each store has roughly 60,000 individual products on its shelves. The
individual products are called stock keeping units (SKUs). About 55,000 of the SKUs come from outside
manufacturers and have bar codes imprinted on the product package. These bar codes are called universal
product codes (UPCs). UPCs are at the same grain as individual SKUs.
At the grocery store, management is concerned with the logistics of ordering, stocking, and selling products
while maximizing profit.
Dimensions:-
A careful grain statement determines the primary dimensionality of the fact table.
It is then often possible to add more dimensions to the basic grain of the fact table, where these
additional dimensions naturally take on only one value under each combination of the primary
dimensions.
If the additional dimension violates the grain by causing additional fact rows to be generated, then the
grain statement must be revised to accommodate this dimension.
In our case study we identify 3 primary dimensions: the date, product, and store dimensions.
Additionally we can add new dimensions, as the promotion under which the product is sold.
In our case study we decide on the following descriptive dimensions: date, product, store, and
promotion.
In addition, we’ll include the POS transaction ticket number as a special dimension (described later)
Page 2 of 14
Date Dimension:
Major Event
Page 3 of 14
Unlike most of our other dimensions, we can build the date dimension table in advance. We may put 5 or 10
years of rows representing days in the table so that we can cover the history we have stored, as well as several
years in the future. Even 10 years’ worth of days is only about 3,650 rows, which is a relatively small dimension
table.
1. Each column in the date dimension table is defined by the particular day that the row represents.
2. The day-of-week column contains the name of the day, such as Monday. This column would be used to
create reports comparing the business on Mondays with Sunday business.
3. The day number in calendar month column starts with 1 at the beginning of each month and runs to 28,
29, 30, or 31, depending on the month. This column is useful for comparing the same day each month.
Similarly, we could have a month number in year (1, ... , 12)
4. The day number in epoch is effectively a Julian day number (that is, a consecutive day number starting at
the beginning of some epoch).
5. For reporting, we would want a month name with values such as January. In addition, a year month
(YYYY-MM) column is useful as a report column header.
6. We likely also will want a quarter number (Q1, ... , Q4), as well as a year quarter, such as as well as a
year quarter, such as 2001-Q4. We would . We would have similar columns for the fiscal periods if they
differ from calendar periods
Store Dimension:
1. The store dimension is the primary geographic dimension in our case study.
2. Each store can be thought of as a location. Because of this, we can roll stores up to any geographic
attribute, such as ZIP code, county, and state in our country. Stores usually also roll up to store districts
and regions.
3. The floor plan type, photo processing type, and finance services type are all short text descriptors that
describe the particular store. These should not be one-character codes but rather should be 10- to 20-
character standardized descriptors that make sense when viewed in a pull-down list or used as a report
row header.
Store Name
Store City
Store County
Store State
Page 4 of 14
Store Zip Code
Store Manager
Store District
Store Region
Product Dimension:
1. The product dimension describes every SKU in the grocery store. While a typical store in our chain may
stock 60,000 SKUs.
2. when we account for different merchandising schemes across the chain and historical products that are
no longer available, our product dimension would have at least 150,000 rows and perhaps as many as a
million rows.
3. The product dimension is almost always sourced from the operational product master file.
4. Most retailers administer their product master files at headquarters and download a subset of the file to
each store’s POS system at frequent intervals.
5. It is headquarters’ responsibility to define the appropriate product master record (and unique SKU
number) for each new UPC created by packaged goods manufacturers.
6. An important function of the product master is to hold the many descriptive attributes of each SKU.
7. The merchandise hierarchy is an important group of attributes. Typically, individual SKUs roll up to
brands. Brands roll up to categories, and categories roll up to departments.
Product Description
Brand Description
Category Description
Department Description
Page 5 of 14
Package Type Description
Package Size
Fat Content
Diet Type
Weight
Storage Type
Shelf Width
Shelf Height
Shelf Depth
Promotion Dimension:
1. The promotion dimension is potentially the most interesting dimension in our schema.
2. The promotion dimension describes the promotion conditions under which a product was sold.
3. Promotion conditions include temporary price reductions, end-aisle displays, newspaper ads, and
coupons.
4. This dimension is often called a causal dimension (as opposed to a casual dimension) because it
describes factors thought to cause a change in product sales.
5. Managers at both headquarters and the stores are interested in determining whether a promotion is
effective or not.
6. From a purely logical point of view, we could record very similar information about the promotions by
separating the four major causal mechanisms (price reductions, ads, displays, and coupons) into four
separate dimensions rather than combining them into one dimension
Promotion Name
Ad Type
Display Type
Coupon Type
Ad Media Name
Page 6 of 14
Display Provider
Promotion Cost
Facts:-
The gross profit is obtained by subtracting the cost dollar amount from the sales dollar amount.
All these facts (sales quantity, sales dollar amount, cost dollar amount and gross profit dollar amount)
are additive across all the dimensions.
Percentages and ratios, such as gross margin, are non additive (gross margin = gross profit / dollar 33
revenue) The numerator and denominator should be stored in the fact table.
The ratio can be calculated in a data access tool for any slice of the fact table by remembering to
calculate the ratio of the sums, not the sum of the ratios. Unit price is also a non additive fact (is a
ratio)
Data Model:-
Decide what business data to model by combining an understanding of the business requirements with
an understanding of the available data.
Management wants to better understand customer purchases as captured by the POS system the
business process we’re going to model is POS retail sales.
This data will allow us to analyze what products are selling in which stores on what days under what
promotional conditions
Key Insights:-
In our case study, the most granular data is an individual line item on a POS transaction.
Rather than representing transaction line item detail in the dimensional model, we could select sales
data rolled up by product and promotion in a store on a day.
A data warehouse almost always demands data expressed at the lowest possible grain of each
dimension not because queries want to see individual low-level rows, but because queries need to cut
through the details in very precise ways.
Providing access to the POS transaction information gives us with a very detailed look at store sales:
To assess whether it’s worthwhile to stock so many individual sizes of certain brands, such as cereal.
To understand how many shoppers took advantage of the 50-cents-off promotion on shampoo.
To determine the impact in terms of decreased sales when a competitive diet soda product was
promoted heavily
Page 7 of 14
GFU is a $5 billion property and casualty insurer that offers automobile, homeowners’, and personal property
insurance.
The primary value chain of an insurance company is seemingly short and simple. The core processes are to issue
policies, collect premium payments, and process claims. The organization is interested in better understanding
the metrics spawned by each of these processes. Users want to analyze detailed transactions relating to the
formulation of policies, as well as transactions generated by claims processing. They want to measure profit over
time by coverage, covered item type (that is, which kinds of houses and automobiles), geographic, demographic,
and sales distribution channel characteristics. Of course, the desire to monitor profit implies that both revenues
and costs can be identified and tracked.
Identify maximum Dimensions and metrics that can satisfy management reporting needs. (2.5 marks)
Interpret each metric to the management so they can realize the associated benefit. (2.5 marks)
According to the International Organization for Standardization (ISO), quality is “the degree to which a set of
inherent characteristics fulfill requirements.” The requirements of a product or process can be categorized or
given a grade that will provide a basis for comparison. The quality is determined by how well something meets
the requirements of its grade
Statistics:
Determining how well products meet grade requirements is done by taking measurements and then interpreting
those measurements. Statistics—the mathematical interpretation of numerical data—are useful when
interpreting large numbers of measurements and are used to determine how well the product meets a
specification when the same product is made repeatedly. Measurements made on samples of the product must
be within control limits—the upper and lower extremes of allowable variation—and it is up to management to
design a process that will consistently produce products between those limits.
High quality is achieved by planning for it rather than by reacting to problems after they are identified.
Standards are chosen and processes are put in place to achieve those standards. The following represents the
quality planning tools available;
Cost-benefit analysis is looking at how much your quality activities will cost versus how much you will
gain from doing them. The costs are easy to measure; the effort and resources it takes to do them are
just like any other task on your schedule. Since quality activities don’t actually produce a product, it
is sometimes harder for people to measure the benefit. The main benefits are less reworking, higher
productivity and efficiency, and more satisfaction from both the team and the customer.
Benchmarking means using the results of quality planning on other projects to set goals for your
own. You might find that the last project in your company had 20% fewer defects than the one
before it. You should want to learn from a project like that and put in practice any of the ideas
they used to make such a great improvement. Benchmarks can give you some reference points
for judging your own project before you even start the work.
Page 8 of 14
Design of experiments is the list of all the kinds of tests you are going to run on your product. It
might list all the kinds of test procedures you’ll do, the approaches you’ll take, and even the tests
themselves. (In the software world, this is called test planning.)
Cost of quality is what you get when you add up the cost of all the prevention and inspection
activities you are going to do on your project. It doesn’t just include the testing. It includes any
time spent writing standards, reviewing documents, meeting to analyze the root causes of
defects, reworking to fix the defects once they’re found by the team: in other words, absolutely
everything you do to ensure quality on the project. Cost of quality can be a good number to
check to determine whether your project is doing well or having trouble
Control charts can be used to define acceptable limits. If some of the functions of a project are
repetitive, statistical process controls can be used to identify trends and keep the processes within
control limits. Part of the planning for controlling the quality of repetitive processes is to
determine what the control limits are and how the process will be sampled.
Cause-and-effect diagrams can help in discovering problems. When control charts indicate an
assignable cause for a variation, it is not always easy to identify the cause of a problem.
Clients provide specifications for the project that must be met for the project to be successful. Recall that
meeting project specifications is one definition of project success. Clients often have expectations that are more
difficult to capture in a written specification. For example, one client will want to be invited to every meeting of
the project and will then select the ones that seem most relevant. Another client will want to be invited only to
project meetings that need client input. Inviting this client to every meeting will cause unnecessary frustration.
Listening to the client and developing an understanding of the expectations that are not easily captured in
specifications is important to meeting those expectations.
Planning for quality is part of the initial planning process. The early scope, budget, and schedule estimates are
used to identify processes, services, or products where the expected grade and quality should be specified. Risk
analysis is used to determine which of the risks to the project could affect quality.
Quality Assurance:
The purpose of quality assurance is to create confidence that the quality plan and controls are working properly.
Time must be allocated to review the original quality plan and compare that plan to how quality is being ensured
during the implementation of the project.
Process Analysis:
The flowcharts of quality processes are compared to the processes followed during actual operations. If the plan
was not followed, the process is analyzed and corrective action taken. The corrective action could be to educate
the people involved on how to follow the quality plan, or it could be to revise the plan.
Page 9 of 14
PAF KIET’s management is eagerly looking for a BI solution as they want to gain insights into students’ data to
improve their academics.
They have bulk of data on paper, Excel and in MIS but it is really difficult for them to analyze the data.
The University has various faculties. Each faculty has several courses and each course has several students. Each
faculty is running various programs and has permanent and visiting faulty members.
For a broader perspective, University management wants a solution that can provide following insights.
Students admissions
Students Pass versus Fail ratio
Top students
Male versus Female ratio
Permanent versus Visiting faculty ratio
For each course, teacher versus students
Number of research paper produced in a year
Number of projects done in a semester per faculty
Draw a Dashboard using DAR approach clearly showing metrics and dimensions as described in the case study
In your own words not from internet, answer these questions in maximum three lines.
a) What is the Objective of Business Intelligence System?
It is known as business intelligence or business intelligence, the set of techniques, strategies, and
technology that through the measurement and subsequent analysis of data, seeks to improve the task
of those responsible for a store and help them achieve their business objectives. The main objectives of
business intelligence are:
1. Collect all possible data about the clientele
Among these data are:
The number of people entering the store at different times of the day. This is done through an advanced
person counter. Tour they make inside the premises. Websites that visit from their mobile phones
through the free Wi-fi provided by the store.
1. Analyze the data collected
A set of numbers is useless if they are not interpreted. Business intelligence technology, in its current
state, provides automated solutions for this.
This way, among other issues, you can find out which are the most and least crowded days and hours or
which are the sectors of the store that go unnoticed in the eyes of customers.
1. Develop strategies to increase profits
With the tools provided by business intelligence and based on the data collected and its subsequent
analysis and interpretation, it is intended to develop strategies to increase profits. It will be looked for:
Increase the percentage of purchase, that is, search that each person who enters the premises becomes
a customer. Loyalty to customers. Get them to return and become regular customers. Reduce expenses.
Eliminate unnecessary actions and issues that report losses to the business and do not provide any
benefit that compensates them.
1. Allow constant control of the person in charge of the store
Under the premise that delegating tasks is important, but your supervision should never be abandoned,
business intelligence gives the person responsible for the store the possibility of having in real time all
the data it needs to make the right decisions.
For example, through a good Shop Managers app, it is possible to know in real time issues such as traffic,
conversion rate, and other relevant KPIs.
Page 10 of 14
An action that provides concrete results All the objectives as mentioned earlier, which previously had to
be reached with a long and complex analysis, or through the simple intuition of the merchants, are now
achievable through business intelligence.
Simply apply these technologies and strategies to start seeing the results and enjoy the benefits
b) What are the three latest trends you understand are important in BI?
Trend #1: There Will Be New Avenues for Data Discovery
Trend #2: More Organizations Will Use BI Platforms
Trend #3: Predictive Business Analytics Will Grow
c) If you have a choice to implement QlikView or QlikSense, which one would you choose and why?
I will choose QlikSense because it can analyze data and generate reports with its own BI
e) At Maxwell Inc., the BI Architect emphasized showing data prominently on Dashboard while minimizing
the effect of borders, gridlines and drop shadows. Another Architect 2 raises concern over this
approach. In your opinion, who is correct and why?
Page 11 of 14
• cable TV systems
Cable companies are now aggressively offering local telephone service and Internet service.
Telecommunications service providers are now selling TV via Internet protocol services,
competing directly against cable for consumers’ entertainment dollars and making the
relationship between the telecom and cable sectors more and more complex.
Ingenuity, innovation, insight and a reasonable approach to spending and investment can help
to move the industry ahead. To drive these, telecommunications service providers will need to
employ cost-effective business intelligence (BI) solutions. This paper discusses how BI, built on
Open Source technologies such as Ingres 2006, can help telecommunications vendors control
costs and improve their bottom line in today’s extremely competitive environment.
Customer Attrition
Acquiring new customers is much more costly than retaining existing ones according to
numerous studies. Customer attrition analysis is an essential step in customer retention.
It involves analysis of data captured during individual customer contacts at various touch
points. For attrition analysis, customer contact data is coupled with other data sources like
billing information. The resultant data set is then associated with customers who have switched
to analyze the possible reasons behind the decision. The results can also be used to improve
the performance of customer touch points.
Customer Affinity
Affinity analysis, or market-basket analysis, reveals linkages between products that are likely to
be purchased together and between product groupings and customer segments. These affinities
can be, at times, extremely difficult to unearth and often business intelligence systems are used
for this purpose. These systems use a technique called ‘association analysis’ for arriving at the
right combination of products and services for a customer or customer segment.
Page 12 of 14
Target Marketing
Marketing to a specific customer group is a natural outcome of customer segmentation. Once
distinct customer segments are identified, BI tools can be used to study the products likely to
be bought by the segment. Often data mining is used to develop predictive models to establish
the buying propensity of a segment towards various existing or new products. Armed with this
information, marketing managers can design specific campaigns targeted at individual segments.
Campaign Analysis
Effective advertising campaigns that consistently deliver on their marketing performance
objectives are the result of skilled advertising personnel, extensive knowledge of the target
market segment as well as excellent understanding of past successes and failures. BI solutions
that bring together data from existing systems that deliver, track, and optimize branding and
direct response marketing campaigns can help telecommunications service providers reduce the
risk and increase the predictability of campaign execution. Campaign analysis is used to analyze
the effectiveness of a marketing or promotion campaign. The effects of a particular campaign on
sales of the promoted product can be tracked using business intelligence solutions. Often the
surge in sales of the promoted product can result in decrease in sales of other related products.
BI tools can also help identify such relationships. The campaign data is stored in a data
warehouse and can be used to predict the effectiveness of similar campaigns in the future.
Cross-Selling
Cross selling can be a major source of selling for a telecommunications company. For effective
cross-selling, existing data can be leveraged using business intelligence solutions to quickly zero
in on new products that may be required by existing customers. These can then be offered to
them during the next contact
Forecasting
To plan their networks, telecommunications service providers perform forecasting that helps operators to
make key investment decisions. These decisions effect all aspects of the business including product
development, launch, advertising, and pricing. Effective forecasting helps to ensure that the company will
make a profit and that capital is invested wisely. BI solutions that use forecast data can help network
planners decide how much equipment to purchase and where to place it to ensure optimum management
of traffic loads.
Budgeting
Data warehousing facilitates analysis of budgeted versus actual expenditure for various
cost heads like promotion campaigns, product development, infrastructure maintenance,
investments, commissions, etc. BI tools can provide drill down capabilities whereby the reasons
for cost overruns can be analyzed in more detail. It can also be used to allocate budgets for the
next financial period. Various activity based costing models can be developed for better cost
control and allocation.
Profitability Analysis
This includes profitability of individual products, product lines, and investments. A
major component of profitability analysis is a thorough analysis of costs incurred during
product development which can be a major factor in reducing the overall profitability of
telecommunications companies.
Corporate Dashboards
Performance measurements like product line profitability, overall development costs, and ROI
can be presented in dashboard reports to top management to facilitate the decision making
process. Also alerts can be triggered if any performance measure reaches a pre-defined threshold
level. These reports can incorporate Telecommunications industry benchmarks provided by
third party researchers.
Statutory Reporting
Telecommunications companies have to provide statutory reports to outside agencies. These
Reports can easily be generated from the Business Intelligence environment.
Page 14 of 14