A Work Project, presented as part of the requirements for the Award of a Master Degree in Management from the NOVA – School of Business and Economics.
PKNOA’s Internationalization Strategy
Appendices
Afonso Marcos Moreira dos Reis Monteiro
Student Number: 18281 / 2933
A Project carried out on the Masters in Management Program, under the supervision of: Prof. Carlos Santos
Table of Contents
Appendix 1:
Additional examples of how Big Data is present in our world ………. 3
Appendix 2:
Steps in a Big Data analytics lifecycle ………. 5
Appendix 3:
PKNOA/DataSonar’s tests and use cases’ results in 2016 ……….. 7
Appendix 4:
Global Competitiveness Index 2015-2016 Rankings ………...…10
Appendix 5:
PKNOA Countries Ranking ……….…14
Appendix 6:
Calculations - number of business problems identified ... 15
Appendix 7:
Graph “Relationship nº problems / investment made” and trendline with R2………... 16
Appendix 8:
Calculations & graph “Relationship nº problems / plans to invest” ………... 17
Appendix 9:
Examples of Smart Meters ……….. 18
Appendix 10:
Appendix 1
Additional examples of how Big Data is present in our world
- Data is growing faster than ever before and by the year 2020, about 1.7 megabytes of new information will be created every second for every human being on the planet;
- By then, our accumulated digital universe of data will grow from 4.4 zettabytes today to around 44 zettabytes, or 44 trillion gigabytes;
- We perform 40,000 search queries every second which makes it 3.5 searches per day and 1.2 trillion searches per year;
- In August 2015, over 1 billion people used Facebook in a single day;
- Facebook users send on average 31.25 million messages and view 2.77 million videos every minute;
- Every minute up to 300 hours of video are uploaded to YouTube alone;
- In 2015, a staggering 1 trillion photos will be taken and billions of them will be shared online. By 2017, nearly 80% of photos will be taken on smart phones;
- In 2015, over 1.4 billion smart phones will be shipped – all packed with sensors capable of collecting all kinds of data, not to mention the data the users create themselves;
- By 2020, we will have over 6.1 billion smartphone users globally (overtaking basic fixed phone subscriptions);
- Within five years there will be over 50 billion smart connected devices in the world, all developed to collect, analyze and share data;
- Google uses distributed computing (performing computing tasks using a network of computers in the cloud) every day to involve about 1,000 computers in answering a single search query, which takes n more than 0.2 seconds to complete;
- The Hadoop (open source software for distributed computing) market is forecast to grow at a compound annual growth rate 58% surpassing $1 billion by 2020;
- Estimates suggest that by better integrating big data, healthcare could save as much as $300 billion a year – that’s equal to reducing costs by $1000 a year for every man, woman, and child;
- The White House has already invested more than $200 million in big data projects;
- For a typical Fortune 1000 company, just a 10% increase in data accessibility will result in more than $65 million additional net income;
- Retailers who leverage the full power of big data could increase their operating margins by as much as 60%;
- 73% of organization have already invested or plan to invest in big data by 2016;
- At the moment less than 0.5% of all data is ever analyzed and used, just imagine the potential here.
from the Forbes article “Big Data: 20 Mind-Boggling Facts Everyone Must Read”
Appendix 2
Steps in a Big Data analytics lifecycle
Data Identification: identifying the datasets required for the analysis project and their sources. The more sources a company can identify, the higher the probabilities of finding hidden patterns and correlations
Data Acquisition & Filtering: where data is gathered from every source identified before, which is then subject to filtering to eliminate corrupt or useless data.
Data Extraction: where disparate data is extracted and transformed into a suitable format for the Big Data solution, because some of the identified data may be in incompatible formats and types.
Data Validation & Cleansing: establishing validation rules and removing invalid data, since sources external from the company are unstructured, without any indication of validity
Data Aggregation & Representation: arriving at a unified view by integrating multiple datasets together, which can be complicated due to different data models and semantics (for example, one datasets may have “surname” while another may have “last name”, although they are the same thing). This stage many times require effort and a significant amount of time.
Data Visualization: to allow non-technical users to understand the insights obtained, data visualization techniques are necessary to graphically transmit the results to support the decision-making process to business users.
Appendix 3
PKNOA/DataSonar’s tests and use cases’ results in 2016
Tests
§ 140.000.000 records per second in reading operations from disk, which was 73 times faster than Cassandra, 33 times faster than Microsoft SQL, 25 times faster than HBase and 47000 times faster than MongoDB.
§ 5.500.000 records per second in writing operations to disk, which was 2.3 times faster than Cassandra, 1.4 times faster than HBase, 140 times faster than Microsoft SQL and 45.8 times faster than MongoDB.
Use Cases in different industries
Ø Financial Services
Client Problem: The client was unable to correctly predict daily activity of 12 different line services of transactional processes, used to create alarms in business activity. This issue led to an increase of false-positive alarms.
DataSonar’s Challenge: Improve the accuracy of the predictive models for each line service, based on 5 years of historical data.
Achievements: DataSonar was able to increase the accuracy of predictive models up to 60% and consequently reduce false-positive alarms.
Ø Insurance
DataSonar’s Challenge: Perform 20 predictive rules and create an output score that indicates fraudulent activity for each claim, under the same business request (200.000 claims).
Achievements: With the same accuracy than the installed solution, DataSonar results showed a performance 746 times faster.
Ø Oil & Gas
Client Problem: The client intended to improve the performance of extracting seismic traces for viewing images derived from SEG-Y files, with an approximate size of 400GB.
DataSonar’s Challenge: Ability to process the images faster than the solution OpenDTect and PETREL.
Achievements:
DataSonar Petrel OpenDTect
Inline 500ms 1s 30s
Crossline 55s 1min 1 min
Z-shape 475ms 108min 100min
Performance comparison in the Oil & Gas use case
Ø Distribution and Logistics
DataSonar’s Challenge: (1) real-time monitoring of up to 4 million data point per day, with different sampling; (2) query historical data of approximately 1 billion data points that corresponds to 7 months of activity.
Achievements: (1) development of a real-time and historic dashboard with search box; (2) high levels of performance, querying approximately 1 billion data point under 1 second.
Ø Utilities
Client’s Problem: Excessive response time to complaints regarding the productivity of domestic solar panel energy, due to the multiple complex data sources needed to create the report.
DataSonar’s Challenge: Integrate 11 different databases in our platform, which have different data formats, in order to create a real-time client report.
Appendix 4
Global Competitiveness Index 2015-2016
Appendix 5
Appendix 6
Calculations – number of business problems identified
1ST STEP: calculate the total number of problems identified per industry Example 1
34 people in the Manufacturing
industry identified Enhanced Customer Experience as a Big Data problem:
52% x 65 = 34
Example 2
18 people in the Services industry identified Improved Risk Management as a Big Data Problem:
29% x 62 = 18
2ND STEP: calculate for each industry the average number of problems each person identified
Example 1
On average, each person in the Government industry identified 2,77 Big Data problems:
97 / 35 =2,77
Example 2
Appendix 7
Appendix 8
Appendix 9
Examples of Smart Meters
from the Radiation Shield website
http://smartmeterradiationprotection.com/author/admin/page/2/
from the The Telegraph article “The £350 smart meter glitch that will last until 2017”
Appendix 10
Example of a Smart Grid
