Abstract ‘ For companies maintaining direct contact with large numbers of customers however, a growing number application like e-commerce, call centre support create a new data management challenge: that is effective way of integrating enterprise applications in real time. To learn and study from the past and forecast the future, many companies or organizations are adopting Business Intelligence (BI) tools and systems. Through BI concepts Companies or organizations have understood the importance of enforcing achievements of the goals defined by their business strategies. This paper explores the concepts of BI, its components, emergence of BI and BI benefits, factors influencing BI, technology requirements and various BI techniques.
Keywords ‘ Business Intelligence (BI), BI components, BI techniques, BI benefits, BI process, BI Tools, Artificial Intelligence (AI), On-Line Analytical Processing (OLAP)
I. INTRODUCTION
Business Intelligence (BI) can be defined as the skills, processes, technology, applications and practices we use to support decision making decisions in organisations. It is a broad category of applications and technologies which is used for gathering, collecting, storing, enlacing and providing access to the collected data to help people in making better business decisions.
BI is a system that gathers, integrates, analyses, concludes and presents business information to support better business decision making. It is an environment in which business users receive information that is correct, reliable, protected, consistent, understandable, easily manipulated and timely. It facilitates more informed decision making.
In order to understand more easily in a single line we can say BI makes possible to convert ‘Data’ into ‘Information’.
BI technologies provides historical, authentic, current, and predictive views of business operations. Its Common functions are reporting, analytical processing, analysis, text mining, employee performance management, setting benchmark, data mining, and predictive analysis. In order to learn from past and forecast the future; now companies use BI to improve decision making and identify new business opportunities. Let us take an example of Restaurant chains in which they use BI to make strategic decisions, such as what new products to add or remove to their menus, which dishes to remove etc. They also use BI for other purpose such as renegotiating contracts with food suppliers and identifying opportunities to improve inefficient processes because restaurant chains are so operations-driven and BI helping them run their businesses. Business Intelligence often aims to support better business decision-making. Thus it can also be called a Decision (DSS). Though the term business intelligence is often used as a synonym for competitive intelligence, as they both support decision making. For using Business intelligence, analysts must understand what viewers are interested in and how business is run, but they must also have the technical and managing skills to solve complex queries, design intuitive reports, and so on [1]
II. RELATING AI (ARTIFICIAL INTELLIGENCE) TO BI
Artificial intelligence (AI) is technology and a part of computer science that studies and develops intelligent, smart machines and software programs. Major AI researchers and textbooks define it as the study and design of intelligent agents, where an agent is a system that perceives its environment and takes actions that enhance and maximize its chances of success whereas BI is a set of processes, theories, methodologies, architectures, and technologies that transform primary data into useful information for business purposes. BI can handle huge amount of information which help to identify and develop new business opportunities, which helps in implementing an effective strategy that can provide a competitive market advantage and long-term consistency.
III. COMPONENTS OF BI
OLAP (On-line analytical processing): It also called international processing analysis. OLAP performs functions like slice, dice, rollup and drill down. Slice and dice allows the navigation of dimensions such as time or hierarchies. Rollup provides data from lower to upper hierarchies and Drill down (roll down) is opposite to Rollup. OLAP provides multidimensional, summarized views of data which is used for reporting, modelling, analysis, and planning for optimizing the business and millions of records can be accessed. It provides the long term informational function e.g. decision support and the design of this is star or snowflake. OLAP sever can be ROLAP which uses relational DBMS to store and manage warehouse data MOLAP which supports multidimensional uses of data to data queue structure and last one is HOLAP which uses the combination of both benefiting from the greater scalability of ROLAP and faster computation of MOLAP. OLAP techniques and tools can be used to work with data warehouses. Other BI tools are used to store and analyse data, such as mining and warehousing; decision support systems and forecasting; data sources; document warehouses and document management; knowledge management; information visualization, and dash boarding; management or geographic information systems; Trend Analysis; Software as a Service.
Advanced Analytics: It is referred as data mining, forecasting or predictive analytics.
Data mining: It refers to an extraction of interesting information or patterns from data in large databases.
In order to extract useful patterns we need to go through some steps: Data cleaning which is used to remove noise and inconsistent data; Integration where multiple data sources may be combined; Selection where data relevant to analysis are retrieved from database; Data transformation where data is transformed into appropriate format to perform summary or aggregation and finally we reach that step where intelligent methods are applied to extract data patterns.
Corporate Performance Management (Portals, Scorecards, and Dashboards): This category is usually for those which provide a container for several things to integrate so that it tells a story. For instance, a balanced scorecard that displays port lets for financial metrics combined with say organizational learning and growth metrics.
Real time BI: It allows for the real time distribution of information through email, messaging Systems and interactive displays.
Data Warehouse and data marts: Data Warehouse is a repository of information collected from multiple sources, store under unified schema and usually resides at single sites. It is subject oriented; integrated; time-variant and non-volatile collection of data. A data warehouse can be used to analyse a particular subject area and can integrate data from multiple data sources. It stores Historical data and once data is in the data warehouse, it will not change. The data warehouse supports the physical propagation of data by handling the numerous enterprise records for Cleansing, integration, selection, aggregation and query tasks. It contains historical data, live data, not snapshots. Data sources can be operational, historical or external data like from market Research companies or data from the net or information from the already existing data warehouse.
The difference between the two respective terms is Data warehouse collects all the information about subjects spanning the entire organization. It provides corporate wide data organization whereas data marts contain a subset of corporate wide data that is value to specific group of users. Scope is confined to specific selected subjects.
Data Sources: Data sources can be operational databases, historical or external data for example, from market research companies or Internet, or information from the already existing data warehouse environment. It can also be relational databases or any other data structure that supports the line of business applications. They also can reside on many other different platforms and can contain structured information, such as tables, piecharts, spreadsheets, or unstructured information, such as pictures, plaintext and other multimedia information.[2]
IV. TECHNIQUES OF BI
Characterization and Discrimination: It is summarizing the data of the class under study for the target class for ex. To study the characteristics of software products to sale increase by 10% in the last year whereas discrimination is the comparison of target class with one order set of comparative class for the comparing. For ex. User may like to compare the general features of software product to sale increase by 10% in the last year with those to sale decrease by 30% during the same period.
Association analysis: It is a discovery of analysis tools showing attribute value condition that occurs frequently together in a given set of data. It is widely used for market basket or transactional data analysis. E.g. age (x, ’20-29′) ‘ income(x, ’20k-29k’) ‘buys (x, ‘galaxy phone’).
The rule indicates that customer under study,2% (support) are 20-29 years of age and have purchased galaxy. There is 60% confidence that customer in this age and income will purchase a galaxy phone.
Classification: This is the process of finding a set of models that describe data classes to predict the class label is unknown.
Prediction: This model is based on analysis of a set of training data. E.g. decision trees and neural network.
Cluster Analysis: Unlike classification which analyse class label, clustering analysis analyse data without consulting a known class label.
Outlier Analysis: Refers when data doesn’t compile with the general behaviour.
In order to make it more understandable we can go through the Table1. [3]
Table 1: Current BI Techniques
TECHNIQUE DESCRIPTION
Predictive modelling Predict value for a specific data item attribute
Characterization and descriptive data mining Data distribution, dispersion or destruction and exception
Association, causality analysis
(Link Analysis) Identify relationships between attributes
Classification Determine to which class a data item belongs
V. THE BUSINESS INTELLIGENCE CYCLE: THE FOUR LINKED COMPONENTS
How to make our business healthy for this we need to apply BI strategy that should be viewed as the sum of four major processes and these four processes are measure, analyze, plan and improve.
Figure 1: Business Intelligence Improvement Cyc
Measure: This is the first phase of business life cycle. The measure phase is the most important phase for far-reaching process of business intelligence. Let us take an example of blowing up a long thin balloon. As you start blowing up the balloon you can see the part of the balloon closest to your mouth expands first, then expansion extends down the length of the balloon. If you written the words measure, analyze, plan and improve downward the length of the balloon starting to the end, the measure section of the balloon would expand first then the other sections. If you try to blow up any other section of the balloon first before you will find it is not possible. The same goes for the BIIC( ).
In the measure phase, companies "report" the current and historical status of key metrics used to manage their business. We need to answer the questions like can the model be enhanced to produce more targeted results? These measures tell a company the "what" (e.g. what is the position of my business?). Although maximum companies know which fundamental indicators to measure (e.g., sales, profit, etc.), it is very difficult to measure individuals throughout their organization. But with the BI solution, an organization can successfully distribute this information to all the people who affect business inside and outside the enterprise.
Determining problems with data collection and connecting them is a necessary evolution that takes place during the measurement stage. Without knowing the problems, companies cannot move into the next stages of the cycle because to base analysis and planning on a suspect measurement system makes no sense.
Analyze: The second phase is analyzed. During this phase, analysts review and measure the data in new and different ways to see whether they can discover hidden relationships that will help them answer "why" (e.g. Why such errors occurring?). Deciding what is important is based on our understanding, knowledge, expectations and assumptions of what are important to customers and employees. In the evolution of BI, several tools have emerged that simplify the analytical process and make it easy like online analytical processing (OLAP) and data visualization
Plan: After determining some of the reasons "why" things occur in analyze phase, companies and organizations then try to determine the effects on outcomes should they implement changes. After the analysis, the third part of the cycle, the plan phase, starts. In this phase, companies use tools to play "what if" with their data. This segment of the BIIC has been divided into planning, budgeting and future forecasting. By using these tools, we can perform scenarios such as "expected and high measures from the budgeting process" and then combine them with historical past measures and forecasting algorithm formulas to determine potential future outcomes. We can then vary our inputs to see how different courses of action might affect these outcomes.
Improve: The plan phase logically progresses into the fourth stage called improve, or the "how" phase. In this phase, analyst within the company discuss outcomes and potential best solutions to the problems they have uncovered in the earlier stages and then make decisions regarding how to make them better, such as what they can do to positively affect their bottom line. As a result of the improve phase, new areas of measure may be added to the upcoming future "measure" phase to track the progress of decisions made during the last cycle.
In this way, a company’s BIIC is a process of perpetual improvement that keeps moving the company toward perfection. Once the cycle starts, it is hard to stop it and, moreover, once you see its results, you’ll wonder how you could have lived without it. [4]
VI. BI PROCESSES
Organizations have been using business intelligence (BI) from past many years to monitor, report on, analysis, and improve the performance of their business operations. In this we have Three Types of BI Processing: strategic, tactical, and operational.
Strategic BI is used for managing long-term business plans and goals. Executives and senior managers use this. Used for historical data around months to years.
Tactical BI analyses business operations for a period of days, weeks, or months and manages to achieve strategic goals
Operational BI is concerned with managing and optimizing daily business operations. It provides the right information at the right time to the right business users to enable them to react rapidly to solve business problems [7]
VII. VITAL BENEFITS OF HAVING BI SYSTEM?
Some of the benefits of having a Business Intelligence system include the ability to access data in a common format from multiple sources, a way to measure goals to see the status of your organizations and to track customer behaviour in order to improve services and relationships. When company can make decisions based on timely and correct information, the company can improve its performance. BI also expedites. It also helps in improving customer experience, allowing for the time to time and appropriate response to customer problems, issues and priorities.
This software can also help to track specific product sales and distributors to improve supply and production, along with this track external trend to improve processes, market trends to improve an organization’s competitiveness, and marketing policies.
Some of the benefits are listed below:
1. With BI superior tools, now employees can also easily convert their business knowledge via the analytical intelligence to solve many business issues, like increase response rates from sources like direct mail, telephone, e-mail.
2. Discover money-laundering criminal activities.
3. Determine what combinations of products and service lines customers are likely to purchase and when.
4. Set more profitable rates for insurance Premiums.
Some issues still need consideration, though increased efficiency and extraction of clear information from complex data can improve and reduce the need for data analyst and improve revenues & profits, determine the exact percentage return on an organizations investment in Business Intelligence systems can be difficult. It can take some time to see the real world benefits of Business Intelligence software and executives can easily label the venture as a failure when results are not immediate.
Though the demand for Business Intelligence tools is growing at a rapid pace there is still criticism from the marketplace that the programs are to complex and difficult to use. User resistance can be one of the biggest hurdles for suppliers of Business Intelligence providers to overcome and handle. Another potential problem is the tools of Business Intelligence. This application can be far more user friendly than they used to be, the core use of BI is still reporting the data rather than process management, though it’s slowly started to change. Business Intelligence users must be careful not to mistake between business intelligence and business analytics. [5]
VIII. IS BI WORTH THE CHALLENGE?
In the end, integrating, simplifying and gathering data from multiple sources in an organization enables users to depend upon the informational memory to help them with the decision-making process and problem solving techniques to better understand the customers and market behaviour.
It takes patience, understanding and cooperation from both ends, i.e. the data analysts and the IT specialists, to achieve the desired outcome and benefit from Business Intelligence tools. It can be achieved through comprehensive detailed training on how to use and interpret the information obtained by the applications.
Business Intelligence is now-a-days getting spread out in many industries like
1. Banking/Insurance Sector
2. Manufacturing (Refinery/ Chemical/ petrochemical/ Paper & Pulp)
3. Pharmaceutical Companies
4. Automobile industry
5. Telecommunication
IX. THE FUTURE OF BUSINESS INTELLIGENCE
Years ago, A 2009 Gartner paper predicted these developments in the business intelligence market.
1. Due to lack of information, processes, and tools. since 2012, more than 35 percent of the top 5,000 global companies and MNC’s were regularly fail to make insightful decisions about significant beneficial changes in their business, trends and markets.
2. By 2012, business units will control at least 40 percent of the total budget for business intelligence.
3. By 2010, 20 per cent of organizations will have an industry-specific analytic application delivered via software as a service as a standard component of their business intelligence portfolio.
4. In 2009, collaborative decision making will emerge as a new product category that combines social software with business intelligence platform capabilities.
5. By 2012, one-third of analytic applications applied to business processes will be delivered through coarse-grained application mash ups.[6]
CONCLUSION
Business intelligence playing an increasingly critical and vital role in tomorrow’s ever faster, ever more globally, even more competitive business environment. In order to meet all these needs, future BI systems has to be real-time, proactive. We can see a bright future emerging for business intelligence. Powerful, transaction-oriented information systems are now very common in every big sector and major industry, effectively levelling the playing field for corporations and organizations globally. The business Intelligence (BI) has much evolved over the past Decade to rely increasingly on real time sourced data. The BI systems auto-initiate actions to systems based on rules and context to support several businesses Processes. Enterprises today demand quick and steady. Now days, not just business analysis has become essential to be done, but also actions in response. So that analysis of results can be performed and instantaneously changes parameters of business processes.
REFERENCES
1. Jayanthi Ranjan, ‘The overview of BUSINESS INTELLIGENCE’, Issues in JATIT, 2005-2009.
2. John Lloyd,’ Identifying Key Components of Business Intelligence Systems and Their Role in Managerial Decision making’, issues in Interdisciplinary Studies Program, February 2009.
3. Rina Fitriana, ‘BUSINESS INTELLIGENCE Techniques, issues in Journal of Theoretical and Applied Information Technology,2005-2009 JATIT
4. Gangadharan.G.R. and Swamy, N., Sundaravalli (2004) ‘Business Intelligence Systems: Designed Implementation Strategies’, Proceedings of26th International Conference on Information Technology Interfaces, Cavtat, Croatia.
5. Marcus Gibson,’ 295 Evaluating the Intangible Benefits of Business Intelligence: Review & Research Agenda, February 2009
6. Eriyatno, Taufik Djatna,’Progress in Business Intelligence System research : A literature Review’, issues in International Journal of Basic & Applied Sciences IJBAS-IJENS ,Vol: 11 No: 03
7. Colin White,’ The Next Generation of Business Intelligence: Operational BI’,issue in BI Research, July2006 sponsored by Sybase
8. Barbara Dinter, Anja Lorenz,’ Social Business Intelligence a Literature Review and Research Agenda