Business Analytics and Data Visualization Decision Support Systems Chattrakul Sombattheera
Agenda • • Business Analytics (BA): Overview Online Analytical Processing (OLAP) Reports and Queries Multidimensionality Advanced Business Analytics Data Visualization Geographic Information Systems (GIS) Real-time Business Intelligence, Automated Decision Support (ADS), and Competitive Intelligence • Business Analytics and the Web: Web Intelligence and Web Analytics • Usage, Benefits, and Success of Business Analytics
Business Analytics (BA): Overview • • • Business analytics (BA) is a broad category of applications and techniques for gathering, storing, analyzing, and providing access to data to help enterprise users make better business and strategic decisions. BA is also known as analytical processing, BI tools, BI applications, and just BI. BI is becoming a major tool for most medium and large corporations. Pizza Hut knows what kind of pizzas customers order, what kind of coupons they usually use, and how much customers spend in a given time period. Marketing managers can run this information through a BI analysis that forecasts, for example, the probability of a customer’s next order. The company then uses this information to determine marketing strategies to influence the customer to buy more pizzas without spending more on that marketing strategy than it has to. Example: An analytic application used for a loan application might: – – Calculate a credit-worthiness score Automatically accept or deny the loan application Select the loan limit Select which credit card product (interest rate, payment terms, etc. ) to issue to this application or which other type of loan to approve.
Tools and Techniques of BA Micro. Strategy’s classification of BA tools: • Enterprise reporting: highly formatted static reports destined for broad distribution. They are pixel-perfect report formats for operational reporting and dashboards. • Cube analysis: provide simple OLAP multidimensional slice-and-dice analytical capabilities to business managers in a limited-range environment. • Ad hoc querying and analysis: power users can query a database for any answer, slice-and-dice the entire database, and drill down to the lowest level of transactional information. • Statistical analysis and data mining: perform predictive analysis or to discover the cause-and-effect correlation between two metrics. • Report delivery and alerting: proactively send full reports or alerts to large user populations (internal and external), based on subscription, schedules, or threshold events in the database.
Tools and Techniques of BA SAP’s classification of strategic enterprise management: operational, managerial, and strategic. SAP offers three levels of support: • Operational: SAP ERP mainly supports transaction processing on the operational level. • Managerial: middle managers can use SAP R/3 to access all reports, arranged by functional areas. Managers can make queries and drill down. • Strategic: SAP offers products under the title SAP SEM (Strategic Enterprise Management), which includes BA.
Tools and Techniques of BA Executive information and support systems. Many BI activities evolved from two tools: • An executive information system (EIS) is a computer based system that serves the information needs of top executives. It provides rapid access to timely and relevant information, to aid in monitoring an organization’s performance by directly accessing management reports and to improve managerial growth and learning. • An EIS is very user friendly, is supported by graphics, and provides the capabilities of exception reporting (i. e. , reporting only the results that deviate from a set standard) and drill down (i. e. , investigating information in increasing details). • Executive support systems (ESS) is a comprehensive support system that goes beyond EIS to include analysis support, communications, office automation, and intelligence support.
Capabilities of EIS/ESS • • Drill-down: the ability to go to additional details at one or several levels. It can be done through a series of menus or by direct queries (using intelligent agents and natural language processing). Critical success factors (CSF): The factors most critical for the success of business. These can be organizational, industry, departmental, etc. Key performance indicators (KPI): the specific measure of each CSF. Status reports: the latest data available on KPI or some other metric, ideally in real-time. Trend analysis: Short-, medium-, and long-term trend of KPI or metrics, projected using forecasting methods. Ad hoc analysis: Analysis made at any time and with any desired factors and relationships. Exception reporting: Using reports that highlight deviation larger than certain threshold. Reports may include only deviations. Slicing and dicing: Rearranging data so that they can be viewed from different perspectives.
Online Analytical Processing (OLAP) • Online analytical processing (OLAP) refers to a variety of activities usually performed by end users in online systems. • Such activities are generating and answering queries, requesting ad hoc reports and graphs and executing them, conducting traditional or modern statistical analyses, and building visual presentations. • Essentially, OLAP products provide modeling, analysis, and visualization capabilities to large data sets, either to database management systems (DBMS) or, more often, data warehouse systems, and they also provide a multidimensional conceptual view of the data. • OLAP differs from online transaction processing (OLTP) that it concentrates on processing repetitive transaction in large quantities and conducting simple manipulations, while OLAP involves examining many data items (frequently up to millions) in complex relationships. Furthermore, users can ask specific, open-ended questions.
Types of OLAP • Types of OLAP: – Multidimensional OLAP: OLAPs that are implemented via a specialized multidimensional databases and summarize transactions into multidimensional views ahead of time. – Relational OLAP: OLAPs that are implemented on top of an existing relational database. – Database OLAP: RDBMS that is designed to host OLAP structures and perform OLAP calculations. – Web OLAP: OLAP data that is accessible from a Web browser. – Desktop OLAP: low-priced, simple OLAP tools that perform local multidimensional analysis and presentation of data downloaded to client machines from relational or multidimensional databases.
Reports and Queries • • • Routine reports: generated automatically and distributed periodically to subscribers on mailing lists. Examples include weekly sales figures, units produced each day and each week, and monthly hours worked. Ad Hoc (or on-demand) reports: created for a specific user whenever needed. An example would be to provide a list of all customers who purchased a company’s products for more than $5, 000 each during January 2006. Examples of Vendors’ Products for Reporting: – – – • Report delivery and alerting: Micro. Strategy offers – – • Business Object’s Crystal Reports, Micro. Strategy Cognos 8 Business Intelligence Hyperion Microsoft’s Report Builder Report distribution through any touch points Self-subscription as well as administrator-based distribution Delivery on-demand, on-schedule, or on-event, Automatic content personalization. Ad Hoc queries can be done by SQL.
Multidimensionality
Advanced Business Analytics
Data Visualization
Geographic Information Systems (GIS)
Real-time BI, Automated Decision Support (ADS), and Competitive Intelligence • • Real-time application are needed due to urgencies: to control (by changing light intervals) traffic flow, to monitor the sea level sensors, etc. Since 2003, BI software tends to produce real-time analysis and decision making tools for operational and tactical personnel--who generally deal with the short-term aspects of running an organization—so that they can use new BA tools and up-to-theminute results to make decision. The proliferation of ADS and business rules management, for example, creates pressure to implement more automated business processes that can best implemented in a real-time data warehouse environment. When processes that require instantaneous updates are needed for answering analytical questions, a realtime response is necessary. To achieve real-time business analysis, real-time data warehouses need to be updated very frequently, not just weekly or monthly. Real-time BA can instantaneously identify, for example, customers buying patterns based on store displays, and recommend immediate changes for placement or the display itself. Other applications include call-center support, fraud detection, revenue management, transportation, and many financial-related transactions. However, all data do not need to be updated continuously, etc. periodic reports.
Automated Decision Support (ADS) • • • Automated decision support (ADS) systems are rule-based systems that provide solutions to repetitive managerial problems. They are also known as enterprise decision management (EDM) systems. ADS systems are close to BA and BI that some BA models are used to create and/or operate the business rules, which can be used to trigger the automatic decision that can be a part of BI applications (performance monitoring and analysis systems). Automating the decision making process is usually achieved by encapsulating business user expertise in a set of business rules that are embedded in a rule-driven workflow (or other action-oriented) engine (which can be part of expert or other intelligent systems). A signal will be passed to the rule engine for evaluation against associated business rules, which determine what action needs to be taken. The four types of business rules: – Analysis rules are used to calculate performance metrics from detailed business transaction data. – Context rules enable performance metrics to be tied to business goals and forecasts. – Exception rules specify thresholds in the metrics and actions (e. g. send alerts) need to be perform when the metrics exceed defined thresholds. – Action rules are a set of manual decision making processes that can be implemented in a rule engine.
Competitive Intelligence • Many companies continuously monitor the activities of their competitors to acquire competitive intelligence. • This information drives business performance by increasing market knowledge, improving knowledge management, and raising the quality of strategic planning. • A sporting goods company found an activist group planning a demonstration and boycott months in advance, enable the company to implement a counter strategy. • Within days of launch, a software firm found dissatisfaction with specific product features, enabling the technicians to write a patch that fixed the problem within days instead of the months normally required to obtain customer feedback and implement software fix. • Barnesandnoble. com used “e-spionage” (Rivalwatch. com) to keep track of competitors in its specialist professional and educational book market by monitoring the price changes from their websites.
Business Analytics and the Web: Web Intelligence and Web Analytics • Most BA applications are related to the Web. Software vendors provide Web tools that connect a data warehouse with e-commerce ordering and cataloging systems (RETSbond, Hitachi’s e-Commerce suite tool, etc. ) • Data Warehousing and analytics and decision support vendors are integrating their products with Web technologies and e-Commerce, and they are also creating new ones for the same purpose (Web Intelligence, Appsource, Decision Edge.
Web Analytics • Web analytics is the application of BA activities to Web-based processes, including e-commerce. This term is used to describe the application of BA to Web sites. The tools and methods are highly visual in nature. • Clickstream Analysis is an implementation of Web analytics. It refers to the analysis of data that occur inside the Web environment; the data, known as clickstream data, provide a trail of the user’s activities and the user’s browsing patterns; which sites are visited, which pages are accessed, how long is spent on the site, etc. • By analyzing and interpreting these data a firm can, for example, find the effectiveness of promotions and determine which products and ads attracts the most attention. • As Clickstream operations increase, the amount of data to process grows exponentially, and scalability issues become critical for Web analytics. • By using sorting and aggregating techniques, one can process 1 billion records per day for a Web data warehouse.
Usage, Benefits, and Success of Business Analytics • Almost all managers and executives can use BA systems, but some find the tools too complicated to use (not trained properly). • Most businesses want a greater percentage of the enterprise to leverage analytics, but most of the challenges related to technology adoption involve culture, people, and processes. • A critical issue is to align BA application to the business needs.
Success and Usability of BA • Organization that have successfully implemented and used analytic applications have realized returns ranging from 17 percent to more than 2000 percent (IDC 2000), with a median ROI of 122 percent. • However, more than half of BI projects fail. • BI activities should be regarded as a constantly evolving strategy, vision and architecture that continuously seeks to align an organization’s operations and direction with strategic business goals. • BI tools can be used to identify inflated invoices, embezzlement, customer impersonation, and similar offenses. • The estimate of total fraud in UK is almost US$30 billion. Fraud committed by employees causes median losses of $60, 000, and fraud committed by managers or executives causes median losses of $250, 000. • When managers and employees conspire, the median loss rises to $500, 000.
Why BI/BA Projects Fail • Failure to recognize. BI projects are different from typical standalone applications that they are cross-organizational. • Unengaged or weak business sponsors. • Unavailable or unwilling business representatives from the functional areas. • Lack of Skilled (or available) staff, or suboptimal staff utilization. • No software release concept (i. e. , no iterative development method). • No work breakdown structure (i. e. , no methodology). • No business analysis or standardization activities. • No appreciation of the negative impact of “dirty data” on business profitability. • No understanding of the necessity for and the use of metadata. • Too much reliance on disparate methods and tools.
System Development and the Need for Integration • Developing and effective BI decision support application can be fairly complex. • Integration, whether of applications, data sources, or even development environment, is a major CSF for BI. • Most BI vendors (Oracle, Business. Objects, Micro. Strategy, IBM, and Microsoft) offer highly integrated collections of applications, including connection to ERP and CRM. • OLAP can be integrated and the output can be analyzed by the neuron network. • Most BI vendors provide for application integration, usually Web enabled.
Скачать презентацию Business Analytics and Data Visualization Decision Support Systems
fe7c24e0dd842ee48cf7d7b90212f37e.ppt