Business Intelligence technology implementation and report

Business Intelligence technology implementation and report.
The students are required to explain and fully report and document their work in a fully detailed technical report as described in steps 1-4 below that demonstrate your knowledge in business intelligence technology. The technology implementation should be also presented during your presentation and you should demonstrate the implementation of the following:
1- Design and Create an excel file that manage the data set for their selected case study
2- Create a Visual Dashboard by connecting the designed excel file with the Tableau software
3- Install and configure Data warehouse using Hadoop, see the following website: https://hadoop.apache.org/
4- Download and demonstrate the using of one of the following ETL tools:
- Talend Open Source Data Integrator
- Scriptella
- KETL
- Pentaho Data Integrator - Kettle
- Jaspersoft ETL
- GeoKettle
- CloverETL
- HPCC Systems
- Jedox
- Apatar
See the following site: https://www.datasciencecentral.com/profiles/blogs/10-open-source-etl-tools
5- Download and demonstrate the usage of WEKA software for Data Mining
https://www.cs.waikato.ac.nz/ml/weka/downloading.html

Report

Business Intelligence and Knowledge Management

Introduction

Business intelligence is an internet-oriented approach for evaluating data and offering reliable information to aid the executives, leaders and other business users to reach informed corporate related decisions. Knowledge management is the art of creating, sharing and using knowledge and information of an organization. It is a multidisciplinary approach to attaining corporate goals though employing the best knowledge. BI uses software and services to change data into dependable information that advises an organization’s strategic and tactical corporate decisions. The rail road is among the hallmarks of industrial revolution but has largely been left behind by the technologies of the internet era. For this study, the discussion is meant to use BI and KM tools to explore the story of Siemens on how the company used Teradata to leverage the sensor-data analytic and predictive maintenance aimed at reducing train failures.

Task 1: Business Intelligence Group Report

Demonstration of business intelligence

In the case study the “Internet of Trains,” it is evident how the train operator in the world is expected to work miracles to maintain their timely services and never be late. To meet this demand, the operators have moved from the reactive maintenance to the predictive maintenance that is seen to be cost effective and condition-based. The train transportation has adapted business intelligence in the industry operations to be sure to meet the current demands in the internet and technology oriented business platform. The internet of train case study implies how Siemens needs to use and re-use its available data to create an internet of trains (Bates, 2015). In this process, the project will be analyzing the sensor data in real time where they can react promptly and ensure client transport service is not interrupted. The case study highlights how the train service can use business intelligence to increase up-time and avoiding an unplanned downtime. While employing business intelligence, the train service can predict incidents earlier in time and clients can react accordingly. BI in action can help clients get more millage from fewer trains. With predictive maintenance, customers will be able to employs their assets better while reducing the costs.

Descriptive analytics

Descriptive analytics employs data aggregation and data mining approaches to give knowledge regarding the past. It is a technique that provides information regarding what transpired and is as well a deeper look at the data trying to understand the causes of the events and behaviors. Most statistics people use fall in the descriptive analytics section as the underlying data is a total or aggregate of a reviewed column of data to where basic math is used. For the case study, a data-based functionality is necessary for Siemens for an effective maintenance program and satisfaction of the available target clients. It appears that in this era of technology, reactive maintenance and regular maintenance with visual monitoring and planned exchange of elements are no longer sufficient. This is the reason Siemens is shifting to a cost-oriented and condition oriented predictive maintenance. To ensure the approach is commercially sustainable, Siemens has to use and re-use the existing data and create an internet of rains (Loshin, 2013). The data sensors will be analyzed in real time meaning they can react rapidly and ensure the customer transport service isn’t interfered with. In the UK, Siemens did a pilot project with a huge train operator where the small data sets of a million sensor-log data were analyzed. The project employed Teradata Aster Discovery Platforms exceptional range analytic techniques to assess the integrated data from various views. The variable that aided in predicting the problems were highlighted while the elements that caused the technical issues or failures of the other elements were measured. An Aster nPath component was used to categories the various sensors showing normal, high, low, and values and tracked the changes.

Predictive analytics

Predictive analytics has focus on the ability to predict the future. Such analytics are usually based on understanding the future and it provide businesses with actionable views based on data. Predictive analytics always give approximate regarding the likelihood of a future result. It is critical to acknowledge that no statistical algorithm can determine the future with 100 certainties. Businesses always employ these statistics to predict what might take place in the coming years. It is because the base of this analytics is founded on potentials. In the case study, Siemens train operators employs the Valero E key components that are frequently evaluated by Siemens. Trains assessing abdominal patterns are dispatched for an inspection aimed at deterring the failures on the track. Such an approach helps the operator to keep their services reliable. The approach helps determine the delays and allows the trains to compete with flights on different routes (Marr, 2017). Data mining technique has been used in the UK where Siemens conducted a pilot project with one operator in the regional routes. Text analysis and mining and sentiment analysis were also applied in the internet of train’s case study. During the validation process, the results from the test data were compared against the total data sets. These showed a higher level of accuracy and proved that the sensor-data analysis makes it possible not to predict the engine failures but launch reactions earlier enough to prevent them. The predictive analytics in the case study improved time though reduction of un-planned downtime, and extension of maintenance intervals because the risks are understood.

The labor costs were reduced and the fix-rates were improved. The case study has not indicated any possible web analytics for Siemens. It appears all the data mining and analysis is currently being done via the offline racks because no website of web data has been given in the case. However, the company appreciates the importance of digitization and its possible in the near future, there will be application of web analytics and eventually web mining will be used in the internet of rains. Siemens has a major focus on digitization and the company acknowledges that digital twin to their physical good is important to give better value to customers. There are a variety of trains and infrastructure components including automation and power systems, railway signaling and control systems that forms a part of the social analytics of the case. The social analytics is also seen where Siemens high speed train Valero E component is continuously monitored by the company (Sabherwal & Becerra-Fernandez, 2009).

Prescriptive analysis

Prescriptive analysis is a case where the users are allowed to prescribe the different potential actions to and guide them through a solution. Such analytics are all regarding giving advice and they make efforts to quantify the effects if the future decisions. Such decisions are aimed at advising the possible outcome prior to making the decisions. Prescriptive analytics always predict what will occur and also why the same will take place and give recommendations on the actions that will take note of the forecasts. For the case of the internet of trains, the users can be allowed to contribute to making the services and the operations better. In this case, the customers are the end users and they are the key stakeholders who are to be satisfied. The efforts to make the train operations on time and to eliminate the delay cases are all meant to satisfy the end users who are the customers or passengers using the trains (Sherman, 2015). For prescriptive analytics, Siemens can seek information from the customers who will potentially give variables of data that may be used as prescriptions to make the services and the operations better. The customers can also spot the areas where there needs to be adjusted and will potentially enlighten Siemens on these areas. The suggestions and opinion from the customers can be used as prescriptions for the company aiming at making the services and operations better for better satisfaction in the future.

The internet of trains has also acknowledged the importance of automated services and this can be linked to the automated decision systems and expert systems. This is where in the future, most of the services and operations in the company will be founded on automated service deliveries and auto-responses including auto-decisions. In this perspective, knowledge management will be applied because the company can use the knowledge collected from the users and from the other stakeholders and keep it for future references. Such knowledge will be kept as prescription for future references and the stakeholders will be part of the collaborative systems giving prescriptions for future corporate success.

Big Data and Future Directions

Big data is usually a complex process of examining large and varied data sets to bring out the hidden information and the unknown correlations and market trends. The company can opt to use the different software and systems to manage big data for the company or seek the services of the cloud computing services. Such services can assure the train operators of real-time data management and low chances of failures and delays. Dig data and analytics will give the train operators opportunity to manage their data and schedule maintenance for their trains and other services (Geng & Wiley, 2017). The systems of software used for big data and analytics in the company will set reminders for the company on any scheduled maintenance, and departures and arrivals in real time. This will make it easier for the company to communicate with the customers in real-time and they interactions will make the business better in the future. Also, knowledge management is critical for a business and is considered the art of creating, sharing and using knowledge and information of an organization. This multidisciplinary approach can help organizations in attaining corporate goals though employing the best knowledge. BI uses software and services to change data into dependable information that advises an organization’s strategic and tactical corporate decisions. The rail road is among the hallmarks of industrial revolution but has largely been left behind by the technologies of the internet era

In future, it is recommended the company should employ a wide use of big data and analytics technique in the operations as this will allow opportunities to successfully meet the customer demands. Web analytics can as well be employed in the event big data is employed and the business will have set a platform where the customers and the operators or service providers will be able to interact and communicate in real time. They will be able to share, interact and offer suggestions and solutions to the different problems they might face and result in successful business intelligence oriented business.

Task 2

Introduction

Scriptella is an open source ETL tool written in Java. The primary focus of the tool is simplicity and it does not need a user to learn complicated XML based language to use it. It enables the use of SQL or any other language feasible for the data source to carry out required transformations. This tool does not need any graphical user interface.

Features of Scriptella

This ETL tool has plenty of features as described below.

·      It supports for many data sources in an ETL file.

·      It has many JDBC features i.e. parameters in SQL including file blobs and JDBC escaping.

·      The tool has high performance and uses low memory.

·      It supports evaluated expressions and properties.

·      It executes different transactions.

·      It is a simple tool and requires no hassle-some installation.

·      The tool has easy to run ETL files directly from Java code.

There are several uses of using this ETL tool. These uses include database migration, database creation, cross database ETL operations and automated database schema upgrade.

Using Scriptella

Usage of WEKA software for Data Mining

How to Run Your First Classifier in Weka

Weka helps in making applied machine learning simple and efficient. This GUI tool enables the user to load datasets, run algorithms and experiments with results that are easy to publish. There is a series of steps that can be followed to run first classifier in Weka appropriately. It is crucial that the user has proper knowledge of the work he is doing.

Conclusion

BI can enable organization to be efficient and productive. Also, features such as web analytics can be used in the event big data is employed and the business will have set a platform where the customers and the operators or service providers will be able to interact and communicate in real time. These can help organizations to share, interact and offer suggestions and solutions to the different problems they might face and result in successful business intelligence oriented business.

References

Bates, J. (2015). Thingalytics: Smart Big Data Analytics for the Internet of Things. Software AG.

Geng, H., & John Wiley & Sons. (2017). Internet of things and data analytics handbook. Hoboken: Wiley.

Marr, B. (2017). Data strategy: How to profit from a world of big data, analytics and the internet of things. London, England ; : Kogan Page

Sherman, R. (2015). Business intelligence guidebook: From data integration to analytics. Waltham, MA: Morgan Kaufman.

Loshin, D. (2013). Business intelligence: The savvy manager's guide. Waltham, MA: Morgan Kaufmann.

Sabherwal, R., & Becerra-Fernandez, I. (2009). Business intelligence. Hoboken, N.J: Wiley.

Print this post