IS 2 Long Report Pardeep kumar 1271107

1 | P a g e
Retail Analytics Tools Comparison for FMCG
in Urban city of Pakistan
INDEPENDENT STUDY – II LONG REPORT
By
Pardeep kumar
Fall 2015 / MS (Software Engineering) / Reg No. 1271107
Email: Pardeep_kumar@outlook.com
IS Advisor
M. Ejaz Tayab
MS Program Coordinator
Dr. Husnain Mansoor
Computer Science Department
SZABIST, Karachi Campus
December, 2015

2 | P a g e
Abstract
In Pakistan, retail analytics is new. Big retailers in Karachi, like, Imtiaz super market, Hyperstars,
Dolman Mall, Macro, Ocean mall, Aga Super market, all have data but not in real-time and
insufficient to analyze and deduce strategies. Also, they don’t have analytical tools to work on
customer profiling, inventory insight, customer shopping engagement, etc. for analytics.
On the retail analytics system to practically deduce results. Data comes in Excel format which
was analyzed in a reputable analytical tool, such as, Qlikview 11.0 version, IBM Cognos Insight
noncommercial and Tableau 9.0. Version.
This paper is based on experimental work to prove the importance of Retail Analytics Tools.
Design Dashboards and compared various commercially available analytical tools. We have
tested practically how data is load in these different tools, how they process and analyze data,
designed three different Dashboards and compared the results as well with same data.
We have also worked on the first challenge using Estimate’s Bluetooth based beacons. We have
captured real-time data and used MS Excel Graphs to understand importance of retail analytics.
Keywords: Retail Analytics, Dashboard, Qlik view, IBM Cognos insight, Tableau.

3 | P a g e
Contents
Introduction ..................................................................................................................................................4
Problem Statement:......................................................................................................................................5
Research Methodology.................................................................................................................................5
Research Structure and Tool.........................................................................................................................6
Stage 1...................................................................................................................................................6
Stage 2...................................................................................................................................................6
Research Scope.............................................................................................................................................6
Field of the Invention....................................................................................................................................6
Background and Prior Art..............................................................................................................................6
Comparison Analytics Tool............................................................................................................................7
Data Analytics on Qlikview............................................................................................................................8
Dashboard on the Tableau ..........................................................................................................................19
Dashboard on the IBM Cognos Insight .......................................................................................................28
....................................................................................................................................................................28
Data Analytics on Excel Sheet.....................................................................................................................37
....................................................................................................................................................................39
Compare the Technically These Three Different Tools...............................................................................39
Retail Sooper Market device can a used Estimate Beacons. ......................................................................41
Retail Data Analytics on Display in Graph...................................................................................................42
Conclusion...................................................................................................................................................43
Future Work................................................................................................................................................44
Acknowledgement......................................................................................................................................44
Appendix A –Important contributors..........................................................................................................44
References ..................................................................................................................................................45

4 | P a g e
Introduction
We are live in the competition in the world where manufacturers, distributors, and
retailers they take all the market present people and business community and FMCG supply
chain are dependent on the they have data present in the Databases and look for retail
Analytics that we are weekly shows[1] in the meeting showing the Dashboard for here
concern is that here data is not present in the not in the organize form first problem face here
is that data must be filter wise and data must come from multiple source that must be in
uniform then we are able to design the Dashboard for higher management then we are also
look into the issue full fill the requirement of the consumer every time his need present in
the [2] Shopping Mall or Sooper store in which event or which Season what are product more
demand they need more in the market if we are fail to the not full fill the requirement then
you are lost the customer you are lost the customer trust so very difficult to gain again.
Mapping of the product also important [3] which product where is present then we attract the
customer also. In Pakistan market competition also face the more issue against the product
here is Data Collection also problem not any sooper market any retail analytics tools present
for these kind of Analysis is present. In our research we are focus on the Data gathering and
how we are Retaial Analytics is Present in the organization old data is present how we are
utilize the old data that is also effective. we are used the Sample data applied on the Tools and
also look itno the see the estimote beacon that device used the via Bluetooth connect Android
App then we are get the data apply then in retail analytics.[4]
Where we have worked on multiple commercially available analytical tools, such as, IBM
Cognos, Tableau, Qlikview and did comparison. We have also worked on the first challenge
using Estimate’s beacon system practically to deduce data in real-time and analyze retail
inventory movements.
Today, retailers gets data from various sources which are very different from each other, i.e. they are all
mixed up. One needs to filter and do data mining to get meaningful data from the big data a retailer
gets.

5 | P a g e
Problem Statement:
In the Retial Analytics there are two main challenges;
That we face how we get the Meaningful data and how we are filtered the data that we
are get the data in one uniform.
In our research, we have focused on the second challenge and have worked on the first challenge
where we have used Estimote’s beacons to get data in real time and deduce inventory movement
using MS Excel. Data comes in Excel format which is analyzed in three reputable retail
analytical tool: Qlikview 11.0 version, IBM Cognos Insight none commercialized and Tableau
9.0. Version.
In Pakistan, retail analytics is new. Big retailers, like, Imtiaz super market, Hyperstars, Dolman
Mall customers, Macro, Ocean mall customers, Aga Super market all have data but not complete,
not in real-time and they don’t have much tools to implement retail analytics.[3]
Research Methodology
I compared three different analytical tools, QlikView, IBM Cognos Insight, and Tableau, I have
used same data to compare results from these three different available tools. I have designed
Dashboards as well to compare results outcome for managers and decision makers. And have
explored different features. I have experimented on real-time data using beacons where Estimote
beacons using excel to analyze the acquired inventory movement data in real-time. Our
Methodology will be both qualitative and quantitative research with experiment.[4]

6 | P a g e
Research Structure and Tools used
I have conducted the research study in two stages:
Stage 1
In this stage I have researched on three tools available commercially in Pakistan. I learned how
to use these tools, how to Extract data, how to design Dashboards. I read Tutorials, saw online
videos, read white papers, articles and forums. [2]
Stage 2
I selected Qlikview, IBM Cognos Insight and Tableau. I practically used Data sample from my
office data, designed different Dashboards on these three different tools. I experimented with
different scenarios and analyzed my results. I also used beacon system to acquire retail
inventory data and analyzed it on MS Excel. [3]
Research Scope
Here is scope is that Retaial Analytics how we are get the data here is data is used in
the tool in the market data combined in the one in the platform in the one uniform data
also very difficult in one format data also gathering issue when we proper Dashboard
then we are solve the problems of the organization also that they grow the business also and
satisfied the customer also that is very challenging task also.[5]
Field of the Invention
In Pakistan, retail analytics is new. Big retailers, like, Imtiaz super market, Hyperstars, Dolman
Mall customers, Macro, Ocean mall customers, Aga Super market, have data but not complete,
not in real-time and they don’t have much tools implement for analytics. Here my invention is
the I applied three different tools, Qlikview, IBM Cognos Insight and Tableau. Used Estimote
beacon systems for retail inventory data in real-time. [6]
Background and Prior Art
Data is Available in the different format in the different organization and different Super Mall
but not applied on the Analytical Tools like Qlikview, IBM Cognos Insight and Tableau. Data
not showing in the Dashboard in real time so [7] organizations take very long time to take correct
and effective decision to compete the market and also not able to fulfill customers’ requirements.

7 | P a g e
Comparison Analytics Tool
Feature Qlickview Desktop Tableau Desktop IBMCognosInsight
Desktop
Visualizations Visualizations are
wizard-driven; colors
have to be selected
Visualizations are drag
and drop, and have
vibrant automatically
generated colors.
Graphs and charts very
old school type and
visually flat.
In-memory BI platform Mean tools used the
own memory fast
processing of data for
quick result showing.
Mean tools used the
own fast processing of
data for quick result
showing.
Mean tools used the
own fast processing of
data for quick result
showing. Engine used
the own memory
ETL Process Tools has own ETL
engine to break the
data into single data
structure.
Blending data from the
different sources here
is ETL to not repeat the
data.
Here is ETL used doing
proper reporting also
perform.
Self-service platform User has own ability to
use the tool no need
of any expert the tool
used
User has own ability to
used
User has own ability to
used
OS Only supported
Windows and not
supported Linux and
Mac
Support for Mac,
windows and Linux.
Supported for windows
and Linux
Mac not supported for.
Data Set Large enterprise-wide
deployments with IT
oversight and
governance.
More commonly used
as departmental vs.
enterprise wide BI
solution.
Government data is
complex and
enormous- so are the
challenges facing those
who work with it. Drop
to visualize any
dataset.
Data Source We are get the data
from multiple data
source also.
We are get the data
from multiple data
source also.
We are get the data
from multiple data
source also.
Vendor Qlik Tableau IBM
1st
Release year 1993 2003 2008
Latest release version. Version 11.0 version: 9.1.0 10.2.2

8 | P a g e
Table1. Comparison of three Analytic Tools
Data Analytics on Qlikview
Fig1- Dashboard Complaint Calls Analytics on the Qlikview
In the above Snaps shot shows that Dashboard source of data is ATM Complaint log
Management system which log the Every ATM call log on the system from all over the Pakistan.
Here above Result shows that most complaint comes from Karachi, second most comes from
Lahore. No of calls per region vise is shown. Other bar chart shows that No of Calls per
engineer. Junaid engineer from Karachi region attended most complaints and Usman attended
most calls from Lahore region. Here company can immediately know from where most calls
come and which engineer Performance is better. Also, call Priority can be established, which
calls are high Priority and which calls are low Priority as shown in the chart. [12]

9 | P a g e
Fig2- Dashboard to show Details of the Complaint Calls
Above chart shows that No of Calls per bank and Location. Here we find which complaint calls
from which bank with high ratio. How many ATMs in the Company, Detail report like Banks
name, ATM issue, which engineer was assigned for the calls, at what time, response time,
Resolved on, Duration of calls, Remedy and Type of workdays. So here the Performance
measure the Banks Complaint how quickly we were able to resolve the Calls.[11]
Fig3. Dashboard for Top Ten issues in Complaint Calls.
Above chart shows which issue comes mostly, why it came, mostly what is the reason behind,
which issues are Top ten that available for the organization to improve performance?

10 | P a g e
Fig4. Dashboard for the No of Calls per Month on the Complaint Calls
Above chart shows No of Calls per month, so here we Show data in month wise report, and
branch wise details. Which call mostly comes from which branch to see which branch suffer
most from ATM Problems, how to reduce the Calls and how to perform better to Increase the
Banks confidence.
Fig 5. Dashboard for the Parts Management on the Compliant Calls
Above chart shows Parts management graph. Parts issued for the Compliant Calls, How we are
managing the Parts for the Company. How these limited parts are used for very important
Complaint Calls as these are also very critical for the Bank’s ATMs.

11 | P a g e
Fig 6. Dashboard for the Response Time of the Compliant Calls
Above Charts Shows the Average Response time for the Compliant Calls in days, also in the
Whole Year or in the Month to see Performance. Overall Calls Performance, Average
Response Time. Support Calls management System Running shows Actual Performance and
How we to improve the Response time and improve the Banks reputation and increase Satisfied
Client.[8]

12 | P a g e
Fig 7. Dashboard for the Parts Issue Management on the Engineer Compliant Calls
Above Snapshot Shows which engineer has used Manual Parts and which Engineer used Auto
parts and how effectively the Parts were utilized.
Fig 8. Dashboard for the Periodic Maintenance for the Compliant Calls

13 | P a g e
Above Charts shows the averages, Average Month for the Periodic Maintenance for the
average for Complaint Calls, how calls were managed effectively, the important calls and
Periodic Maintenance and how effectively resources were used they also for the Complaint
Calls.
Fig 9. Qlikview Associates Different Table Structure
Above Snapshot shows how Qlik view Associates different Tables with Associate keys, how
Relationship between Tables were created, how to extract the data and how Dashboard Design
associated with the Table.

14 | P a g e
Fig 10. Qlikview: the Edit, Script and Load Data
Above Snapshot shows how to load the data from the Sql server OLE DB connector, Table
Files, Qlikview Files, Web Files, Field Data and Excel Sheet data. Also, how to edit the Script
for Association Tables.

15 | P a g e
Fig 11. Qlikview: Edit the Script Connection String and what are the Tables Selected
Above Snapshot shows how in the Qlikview Connection string Build with database. How to
create the connection string build to load the data into Qlikview. OLEDB connection build with
SQL server.[7]

16 | P a g e
Fig 12. Qlikview: Edit the Script Load the Table Data in the QlikView
Fig 13. Qlikview: Edit the Script Load to Update Table Data in the QlikView

17 | P a g e
Above Snapshot show here we are required data Standard time requirement so we are the
Change the time format here we are some function apply also change the time format and change
the date format for our requirement accordingly so we apply here by default function also for
our need and also Extract the date also so mix data comes in the table date and time mix the
data so our requirement only date so we are extract the date only.[9]
Fig 14. Qlikview the Edit the Script Load the Joins Table Data in the Qlik View
Above Snapshot shows that here similarly Joins in the table multiple column used from the
multiple table from so we apply here joins concept in the sql server so here also Qlik view joins
load data accordingly.[8]

18 | P a g e
Fig 15. Qlikview the Edit the Script Load Multiple Table Data Load in the Qlik View
Above Snapshot shows that how we are multiple table data in the Load accordingly our
requirement we can edit the script modify the Column name and modify the what are needs
multiple table load script very easy in the Qlik view also for convince for the developer that
we want that type of data Load. Here also Syntax checker which check the script syntax so easy
for the developer also for the not making any mistake during the data loading. Here in the Qlik
view also display in web also for the convince for the user also. [10]

19 | P a g e
Dashboard on the Tableau
Fig16. Dashboard on the PS Call logged Management on Tableau
Above Snapshot Shows that here highlight from 2014 Calls and 2015 here Calls also so we
are differentiate the which calls from 2014 year and 2015 year also here is small report also
ATM model and Banks Call is assigned from the other Department or Direct from Banks so
here conclusion report . [11]

20 | P a g e
Fig17. How we are Data Load on the Tableau
Above Snapshot shows that How we are connect the Data, text Data, Excel data , Access,
Statistical Files and other files from server Also Tableau Server, SQL Server , Oracle Server,
My SQL server , Amazon Red shift and other Server Also IBM server , Google Analytics, SAP
, Teradata and other Server Also mostly come server below snapshot shows that also.[12]

21 | P a g e
Fig18. How we are Data Load from the Server on the Tableau
Fig19 Dashboard bar Chart PS Call Assigned Different PS Consultant on the Tableau
Above Chart shows that which calls assigned for him where complain came from which date
issue Escalate which ATM Model issue description which year here the conclusion information

22 | P a g e
for the Quick PS Consultant. For him how manage the Calls very effectively for the Resolution
of the Calls.[5]
Fig20 Dashboard Pie Chart Call is assigned on the Quarter wise on the Tableau
Above chart shows that Call is assigned on the Quarter wise that different calls shows in the
every unique calls shows in the Pie Chart that PS Consultant another view the Calls on the
different Angle also for the also which call look into which PS consultant also for resolution
of the call also.

23 | P a g e
Fig21 Dashboard Tree Map chart Call is assigned on the Quarter wise on the Tableau
Above chart shows that on the Tree Map chart all the on the one Snapshot as well as also for one
tree view and or one Hirechcy of the call for very quick information how effectly feel the
difference that chart also PS consultant more easy and more friendly for that kind of chart one
consolidate information in one Cell or one Tree map chart.

24 | P a g e
Fig22 Dashboard on the Bubble Chart for the PS Consultant on the Tableau
Above Chart shows in the Bubble chart this chart also very unique every different PS
Consultant Bubble chart different Color Shows very easy for the PS Consultant whole
Information display for one consultant same color used very easy for Consultant For whole
Call information Display this chart for the that type of chart not available in the Qlik view
and IBM Cognos Insight that chart also very useful PS Consultant as well also.

25 | P a g e
Fig23 Dashboard on the Line Chart PS Consultant on the Tableau
Above chart Shows that Line Chart and Also very Unique Chart for every full information
display for the every PS Consultant that information display unique Chart Display very unique
information for the PS Consultant display the information every click on Symbol for the
Assigned PS Consultant. Here other Detail also you can here set the data on the Dimension and
Applied formula and Calculation apply filter the data for more Explorer the information here
sheet, workbook, and Story also you are design for the Dashboard.

26 | P a g e
Fig24 Data Loading and Filter the Data Tableau
Above Snapshot shows that How we data load in the Tableau here we have two option live
connect the data and Extract the data in the Tableau so live connect the data we have the
benefit that very less time connect the data where in the live information any change so we are
also change the information also update dashboard available for the user also other Qlik
view or IBM Cognos insight that option lack and very beneficial for the developer and End
user for the developer not repetitive over load the load again and again the data for the End
user beneficial is that Update data present for the and update the new information available for
the user and other here we are doing here filter the data also modification also done here and
other column we are not required so we are here delete the column also and filter the data also
according our requirement and need of the End user also and apply function and change the
name also for the according our need. Apply the Script and query also for the developer for full
fill the requirement. [8]

27 | P a g e
Fig25 Data Loading and Filter different option the Data Tableau
Above Snapshot shows that how we are choose different way filter the data likewise the
wildcard contains, Start with , End with and Exactly with Matches and Include all values and
empty here also Exclude the data other option are apply the condition by filed, by range , by
formula and reset the values that above filter why we are change the data and filter the data
because we are not required to the extra load the to the Tableau and also full fill the requirement
and End user need also for very clean the data see the End user full fill the need. Here other
option are Top by field and by formula also we are also that way also filter the data according
our need. Filter the data best option not extra load and extra memory consume and very difficult
restrict the data on the when we design the Dashboard also so very difficult also. Firstly filter
the data very important for the End user also.

28 | P a g e
Dashboard on the IBM Cognos Insight
Fig26 Dashboard on the Bar chart on the IBM Cognos Insight
Above chart shows that Bar chart that most PS Calls from which Bank and how many total
calls for each bank and also shows is that other filter the data here also mention in the other
Department Also calls from and total of All Calls also come from describe shows in the above
chart that very analyses the Which calls from which Banks or other source so very easy for PS
Consultant.

29 | P a g e
Fig27 Dashboard on the Bar chart for Particular Issue on the IBM Cognos Insight
Above chart Shows that For the Particular issue that most comes from which Banks that issue
comes for high in the range that’s PS Consultant analyze why issue come most from the
Particular Banks come from Most that shows in the Above more sure less Problem come from
the Bank.

30 | P a g e
Fig28 How the Data Load from the IBM Cognos Insight
The Data Source Excel file , Text file, ODBC connection , IBM Cognos Report data and IBM
Cognos Package data that you are used for the here the IBM Cognos insight here very
Limited Data Source for used here we filtered the data which data import you want for your
Used here one option Not available that you are used for the connect the Excel data that you
are want .

31 | P a g e
Fig29 Dashboard on the Bar chart different angle IBM Cognos Insight
Above Snapshot shows that Attribute , ALL Dimension and All import data and also display
Bar chart shows in the different angle very straight line view that most calls from which Banks
other issue you are can display as you want for the your information .

32 | P a g e
Fig30 Dashboard on the Pie chart different angle IBM Cognos Insight
Above chart shows that on the display information Pie chart that information display different
charts facility available whole the same information but in the different for the Viewer for
different look into the feel and explorer the more information as you want for detail information
also calculate And other measure value you can also calculate.

33 | P a g e
Fig31 Dashboard on the Pont chart different angle IBM Cognos Insight
Above chart very unique chart Pont chart that count all calls for the Bank that chart very
unique chart that also very unique chart that Pont chart not available in the Qlik view and
Tableau for the same information display that here we choose that same information display in
the very unique angel for the information that’s Gamble for the chart how we gamble the chart
as well as for the information.

34 | P a g e
Fig32 Dashboard on the Scatter and Bubble chart different angle IBM Cognos Insight
Above Snapshot shows same information in the Scatter and bubble chart that chart not attractive
for the Tableau chart display in the different Color in the Bubble that chart very suitable for the
PS Consultant here IBM Congnos Insight as a Bubble chart but not Suitable for the PS consultant
not attractive that how I compare the chart as well as the Chart same information on the
different tool display and different angel want. Also.

35 | P a g e
Fig33 Dashboard on the Tree chart different angle IBM Cognos Insight
Above chart shows that tree chart and tree chart also shows in the Tableau but in the Tableau
very attractive chart also tree chart so we are compare another chart also that tree chart here in
the IBM Cognos insight not the attractive chart so we are conclude the result is that Tableau
chart more suitable and more attractive chart as compare to IBM Cognos insight here same
information which calls from most which Banks so we are got different information and also
not attractive information.

36 | P a g e
Fig34 How we Import the Data and mapping the data IBM Cognos Insight
Above Snapshot shows that how here we are import the data and also mapping the data Target
the items , Mapping source items mapping column according our need and required the data
here items from the source are dimension and measure in the target cube and also here define
the properties and relationship for the items according our requirement and need of the data
and here mapping also hierarchies of the data required or do not required hierarchies so here the
according our need what are the requires here other option are add Calculated items or clear all
mapping and set the properties also for the relationship of the data. Here other option also
Summary the data what are the identify here that whole cube design the data what are the
dimension and what are the measure values so accordingly we are set Dashboard. Here other
option apply also here define also dimension what values unique or measure the values and filter
also apply and apply the condition according our requirement also so the we here modify the
data also. So here properties summary are Cube, Dimension, Levels, Attributes, Measure
accordingly set the properties also for the need and requirement also.

37 | P a g e
Data Analytics on Excel Sheet
Fig35 Excel graph on the “ISSUES”
Above Chart shows graph on “issue” that come in the software. Issues are opened and closed.
Above graph shows issue which are all open. These data are very helpful for the data Analytics.
0
0.5
1
1.5
2
2.5
15-Sep-15 16-Sep-15 17-Sep-15 18-Sep-15 19-Sep-15 20-Sep-15
Issues

38 | P a g e
Fig36 Dashboard on the Excel Sheet Describe one of the issue
Above chart shows that one of the issue described “Network failure”, which was opened and
closed so here we are history of the issue which are close and open for the resolution we are
known in the depth also very clearly identification the issue so that are issue so we are more
focus on the close the issue so we are great efficient work on this very help for the Monitoring
the software and very close on the issue close and open so very helpful the chart.
Fig37 Dashboard on the Excel Sheet Describe Region wise the issue
Above chart shows that Now we are see the issue on the region wise which region wise issue
come high on the software then we are identify that what are reason behind this why issue
come very clear picture on this that so these are issue comes and so these are come now
0
5
10
15
20
25
30
15-Sep-15 16-Sep-15 17-Sep-15 18-Sep-15 19-Sep-15 20-Sep-15 21-Sep-15
Network Failure
open close
0
1
2
3
4
5
6
7
8
Sindh Punjab KPK
Issues

39 | P a g e
region head headache so how we are decrease the issue from the region wise so we have clear
picture on this also.
Fig38 Dashboard on the Excel Sheet Describe ATM wise the issue
Above chart shows that Now we are issue region wise and in the region which ATM issue
high what are reason behind this that issue come from one of the ATM higher and in the
now focus on the Region head to the engineer what are activity perform and or that Particular
ATM has created the problem for us what are the hardware changes and what are the software
changes required to fix the issue so we are the decrease the issue so here we are clear that we
have the data then we are more focus on that we are clear more clear issue on the hand and
resolve the issue also so we are better perform the issue.
Technical Comparison of the three Different Tools.
Qlikview Very user friendly tool and very nice Design the Dashboard so much popular in the
market and also so many client also So much data selection option very large Design the
Dashboard in the Data Loading here also Script the Data also so much option we are change
the Column name and also we are apply. we are require the data from the Quarter Data so in
the data Monthly data is present so apply the formula for the Quarter data get and weekend
formula apply here weekday data required so we are the apply the formula for get the data
according our requirement so we get our result and Analytics accordingly. Here we are also
change the Column the Name and those column also here delete here that we are not required
0
1
2
3
4
5
6
7
ATM1 ATM2 ATM3 ATM4
Issues

40 | P a g e
from the table to get the data and change the data type also perform according our requirement
.we are required data Standard time requirement so we are the Change the time format here we
are some function apply also change the time format and change the date format for our
requirement accordingly so we apply here by default function also for our need and also
Extract the date also so mix data comes in the table date and time mix the data so our
requirement only date so we are extract the date only. how we are multiple table data in the
Load accordingly our requirement we can edit the script modify the Column name and modify
the what are needs multiple table load script very easy in the Qlik view also for convince for
the developer that we want that type of data Load. Here also Syntax checker which check the
script syntax so easy for the developer also for the not making any mistake during the data
loading. Here in the Qlik view also display in web also for the convince for the user also. IN
the Tableau also how we are choose different way filter the data likewise the wildcard
contains, Start with , End with and Exactly with Matches and Include all values and empty here
also Exclude the data other option are apply the condition by filed, by range , by formula and
reset the values that above filter why we are change the data and filter the data because we are
not required to the extra load the to the Tableau and also full fill the requirement and End user
need also for very clean the data see the End user full fill the need. Here other option are
Top by by field and by formula also we are also that way also filter the data according our need.
Filter the data best option not extra load and extra memory consume and very difficult restrict
the data on the when we design the Dashboard also so very difficult also. Firstly filter the data
very important for the End user also. In the IBM Cognos Insight how here we are import the
data and also mapping the data Target the items , Mapping source items mapping column
according our need and required the data here items from the source are dimension and measure
in the target cube and also here define the properties and relationship for the items according
our requirement and need of the data and here mapping also hierarchies of the data required or
do not required hierarchies so here the according our need what are the requires here other
option are add Calculated items or clear all mapping and set the properties also for the
relationship of the data. Here other option also Summary the data what are the identify here that
whole cube design the data what are the dimension and what are the measure values so
accordingly we are set Dashboard. Here other option apply also here define also dimension
what values unique or measure the values and filter also apply and apply the condition according
our requirement also so the we here modify the data also. So here properties summary are
Cube, Dimension, Levels, Attributes, Measure accordingly set the properties also for the need
and requirement also. how here we are import the data and also mapping the data Target the
items , Mapping source items mapping column according our need and required the data here
items from the source are dimension and measure in the target cube and also here define the
properties and relationship for the items according our requirement and need of the data and
here mapping also hierarchies of the data required or do not required hierarchies so here the
according our need what are the requires here other option are add Calculated items or clear all
mapping and set the properties also for the relationship of the data. Here other option also

41 | P a g e
Summary the data what are the identify here that whole cube design the data what are the
dimension and what are the measure values so accordingly we are set Dashboard. Here other
option apply also here define also dimension what values unique or measure the values and filter
also apply and apply the condition according our requirement also so the we here modify the
data also. So here properties summary are Cube, Dimension, Levels, Attributes, Measure
accordingly set the properties also for the need and requirement also.
Retail Super Market can use Estimate Beacons.
A beacon device system can be used to track the product via Bluetooth, we get location by
using various sensors such motion, humidity and temperature that we are get additional
information also in real time so we get inventory, visibility data for retail Analytics. Estimate
Beacons which location present. Beacons are the signal emitting device that transmits radio
signals in specific distance with required signal strength. Beacons are based on silicon casing and
ARM computer combined with Bluetooth device consisting a small battery and low level
software install in beacons.
Beacons are using for broadcasting small amount of data therefore Bluetooth containing only
257 byte data in each packet through the data cell phones are able to determine the signal
proximity. The other principal is advertisement to the available cell phone device in range its
transmits packet after one second with scanning devices but the problem is devices locked and
unlocked in case of locked device beacons required more power with high frequency therefore
beacons eliminate these devices and preserve the power but during the transmission if active
devices move out of rang the data should be distorted so we can design some other application
through SDK that’s help us to transmit two more packets in each scanning if we set it 490ms
give us two packets increase in frequency and in 330ms give us three packets and in 240ms four
packets give. These signals are transmitted with blinking and effect on a battery life. It’s
complicated to determine the exact position of beacon due to public place different types of
obstacles then calculated by RSSI (receive signal strength indicator) in smart phones but for the
precise position tracking we embedded the map on developing apps.

42 | P a g e
Fig39 Estimate Beacons Device via connect the Android app the device shows in the App
Retail Data Analytics on Display in Graph
0
5
10
15
20
25
1 2 3 4 5 6
Shelf Temperature with Hour

43 | P a g e
Fig40 Chart Display Shelf Temperature Varies in Hour
Above Chart shows that the Temperature varies in Shelf with the Hour going we are
recording the temperature as an experiment with our Bluetooth device “estimote beacon
indoor” device that also temperature varies above chart shows that in one hour temperature
20 Centigrade and 2 hour 19 C and so on the chart shows that how we product preserve that
temperature so much varies in the then accordingly we are product set in the shelf.
Fig41 Shelf Life with Product varies with respect to the Hour varies
Above chart shows that the who person know how much product stay in the shelf with
respect to the hour with different Pepsi id shows that that are duration we are calculate with
Time-in and Time out the product in the Shelf the above chart shows that very clearly in
the above chart. How we are the product set in the shelf according the demand of consumer.
Conclusion
I have studied many tools and after studying them, I have selected three tools for my research
work which are, Qlikview, IBM Cognos Insight and Tableau. I have done technical comparison
and selected these Tools for my experiments. I have built three different Dashboards. With
Same data I have compared working of these tools, I have learned Data Loading types,
understand the method of filtering and data cleanup and which data required for analytics and
which data is not required. These tools require technically different method of Dashboard design
and different scripts to load data and different ways of applying filters. I have concluded that
Tableau and Qlikview are better. I have also worked on Estimote beacons to acquire real-time
0
20
40
60
80
100
120
1 2 3 4 5 6 7
Shelf Life with Hour
Shelf1 Duration in Hour Pepsi ID

44 | P a g e
inventory movement data from retailer shelf and found its importance in terms of item shelf-life,
stock-outs and customer trends.
Future Work
In this research study I studied three Tool that I used but more tools are also available so one can
work similar activities on these tools as well. We can also use more sample data and get other
real-time data. We can involve retailers, like, Dolman mall or macro or Hyperstar, can connect
with their database servers and work on their data for retail analytics.
Acknowledgement
I would like to thanks and acknowledge following domain experts and personnel who have all
help me in completing my IS, without their help it would have been impossible to produce such a
good work:
 Wasi UL Akbar
 Syed Haris Hasani
 Arif Ahmed
Appendix A –Important contributors
Below are some of the domain expert’s names along with their designation who contributed and were
considered and helped me in this research.
Organization Employee Name Designation
Touchpoint pvt ltd Wasi UL Akbar Professional Service Consultant
Touchpoint pvt ltd Syed Haris Hasani Professional Service Consultant
PAF KIET Arif Ahmed Student

45 | P a g e
References
1. Josh Brownlow,Mohamad Zaki,Andy Neely and Florian Urmetezer, “Data Driven Busniess Model
A Blueprint for Innvotation” , 2015
2. Auke Hunneman,Peter C. Verhoef and Laurens M.Sloot, The impact of Consumer Confidence
on Store Satisfaction and Share of wallet formation, 2015
3. Vidya Mani,Saravanan Kesavan and JayaShankar M.Swaminathan ,”Estimating the impact of
understaffing on sales and Profitability in retail stores”, 2015
4. Lee A. Carbonel, Flower Mound, (US);TSZ S. Cheng,Grand Prairie,(US);Jeffrey
L.Edgington,Keller,(US);Pandian MariaDoss and Allen,(US),”Automatic Floor-Level Retail
Operations Decision using Video Analytics” ,2015
5. Nizar Zaarour, Emanuel Melachrinoudis,”Perfromance optimization in retail Busniess using real
–time preductive Analytics” ,2015
6. Dr. Abhilas kumar Pradhan,AAkash A. Kamble,Efficiency “Measurement and Bench Marking: An
Application of Data Envelopment Analysis to Select Multi Brand Retail firms in india” ,2015
7. Gloria Phillips-wren, Lakshami S. Layer, Uday Kulkarani and Thilini Ariyachandara, “Busniess
Analytics in the context of Big Data: A Road Map of research”,2015
8. Hans.w.ittmann, “The Impact of Big Data and Busniess Analytics on Supply Chain Management,
2015
9. Hervais Simo, “ Big Data: Opportunity and Privacy Challenges, 2015.
10. Walter Armbruster and Margaret MacDonell , “ Big Data for Big Problems”, 2015
11. Rama Chandara Rao Meka , Dr.Noorullah Shariff c and Amresh patil, Performing Predictive Data
Analytics in Data Mining Using various Tools
12. Catherine Hack, “Applying Learning Analytics to Smart Learning”, 2015

IS 2 Long Report Pardeep kumar 1271107

More Related Content

What's hot (17)

Similar to IS 2 Long Report Pardeep kumar 1271107 (20)

IS 2 Long Report Pardeep kumar 1271107