Week 10 Business Analytics (Chapter 12) ADM 2372 PDF

ADM 2372 Management Information Systems Mayur Joshi, PhD Assistant Professor of Information Systems [email protected] © Copyright. Mayur Joshi. 2024. and © 2024 John Wiley &...

ADM 2372 Management Information Systems Mayur Joshi, PhD Assistant Professor of Information Systems [email protected] © Copyright. Mayur Joshi. 2024. and © 2024 John Wiley & Sons Canada, Ltd. or the authors All Rights Reserved. No part of this document may be reproduced, stored in a retrieval system or transmitted in any form or b y any means, electronic, mechanical, photocopying, recording or otherwise without prior written permission from the professor. Shar ing course materials without permission or uploading course materials to a content sharing website may be treated as an instance of acad emic fraud as well as copyright infringement. 1  Week 10 Business Analytics (Chapter 12) 2 Agenda 1. Business analytics vs intelligence 2. Managers and Decision Making 3. The Business Analytics Process 4. Descriptive Analytics 5. Predictive Analytics 6. Prescriptive Analytics 7. Presentation Tools 3 Business Intelligence (BI) Business Intelligence (BI) is a broad category of applications, technologies and processes for gathering, storing, accessing and analyzing data to help business users make better decisions (P. 364) 4 Business Analytics (BA) Definition – the process of developing actionable decisions or recommendations for actions based on insights generated from historical data BA vs. BI – Some prefer to say BA, some prefer to say BI, some think the two have the same meaning – In this course we consider them interchangeable (P. 364) 5 Business Analytics (BA) Data Data Tools and Business insight Data Techniques e.g., data from e.g., statistical e.g., Capital One databases, models, machine discovered that “the spreadsheets, social learning most profitable media, etc. techniques, etc. customers charge a large amount but pay their credit cards off slowly” 7 6 Decision-Making Role of Managers Management is a process by which an organization achieves its goals through the use of resources (people, money, materials, and information) These resources are considered to be input Achieving the organization’s goals is the output of the process Managers execute and optimise this process (P 365) Output (goal) Input Management People Money Materials Managers optimise Information the process 7 Decision-Making Process and Phases 8 Decision-Making Process and Phases A Decision is a choice among two or more alternatives that individuals and groups make Decision making is composed of three major phases: – Intelligence: examine the situation and identify and define the problem or opportunity – Design: build a model (an abstraction of reality) for addressing the situation; define relationships between the variables of the model; validate the model using test data – Choice: select a solution using evaluation criteria; implement the solution (P 366) 9 Decision-Making Challenges and Need for IT Decision making is difficult due to the following trends: o Number of alternatives is constantly increasing o Most decisions must be made under time pressure o Increased uncertainty in the decision environment o Often necessary to rapidly access remote information, consult with experts, or conduct a group decision-making session 10 Decision-Making Types of problems needing decision-making Structured decisions deal with routine and repetitive problems for which standard solutions exit – e.g., inventory control – Can be automated Unstructured decisions deal with complex problems for which there is no clear and agreed - upon solution or procedure for arriving at the solution – e.g., choosing a set of R&D projects for next year – Human intuition and judgment play a role – BA cannot make unstructured decisions, but can provide information that assists decision makers (P 368) 11 Decision-Making Nature of decisions Operational: executing specific tasks efficiently and effectively Management: decisions concerning acquiring and using resources efficiently in accomplishing organizational goals Strategic: decisions concerning the long-range goals and policies for growth and resource allocation 12 Decision-Making IS support framework 13 Business Analytics (BA) Process BA includes “getting data in” a data warehouse or data mart and “getting data out” using BA applications (P 369) Data warehouse Getting data in Getting data out Data ETL mart BA Applications Data mart Data mart User (decision maker) 14 Business Analytics (BA) Process Three specific analytics targets: o The development of one or a few related analytics applications o The development of infrastructure to support enterprise- wide analytics o Support for organizational transformation 15 Business Analytics (BA) Process 16 Descriptive Analytics Descriptive analytics summarizes what has happened in the past and enables decision makers to learn from past behaviors 17 Descriptive Analytics Tools: OLAP OLAP (Online Analytical Processing) - aka multidimensional analysis – consists of “slicing and dicing” data in a cube (e.g., sales per region), “drilling down” in the data for greater detail (e.g., sales at the store level), and “rolling up” the data for less detail (e.g., sales at the national level) see Topic 3 for examples https://learn.microsoft.com/en-us/system- center/scsm/olap-cubes-overview?view=sc-sm-2025 18 Descriptive Analytics Tools: Data mining Data mining is – Designed to find what queries and reports do not reveal – Mines (i.e., analyzes) the data to identify previously unknown patterns – Examples Mine (analyze) retail sales data to discover products that people often purchase together (e.g., chips and soda); in which case such products are placed next to each other 19 Descriptive Analytics Tools: DSS Decision support systems (DSS) combine models and data to analyze semi- structured (and A Model is an abstraction or reality some unstructured) problems that involve extensive user DSS involvement Three quantitative models Outcome Model used by a DSS – What-if analysis Variables – Sensitivity analysis – Goal-seeking analysis Note: MS Excel supports all three 20 Descriptive Analytics Tools: DSS DSS Sensitivity analysis: interested Model in “what variables have the Outcome most effect on the outcome” - Variables e.g., is the outcome more affected by age, education, or gender? What-if analysis: checks the impact of a change in a Goal-seeking analysis: finds variable on the outcome the inputs (variable values) – “What would be the impact on necessary to achieve an the bottom line if we have a outcome 20% increase in sales” – “in Excel, changing the value of – “how many customers are a cell that is used in a formula required to increase profits to $5 million” to see the result” 21 Descriptive Analytics Big Picture HISTORICAL DATA Operational data database ETL Data warehouse database Data External data mart Applications for Descriptive analytics: ETL Data - OLAP (see cubes in Topic 3) database mart BI applications - Data mining - DSS Data mart 22 Descriptive Analytics Big Picture – OLAP: Primarily used for quick, interactive analysis of pre-aggregated data, enabling users to slice and dice data across different dimensions to gain insights. – Data Mining: Aims to discover previously unknown patterns and relationships within large datasets by applying complex algorithms like decision trees or neural networks. – Decision Support: A broader framework that includes data warehousing, OLAP, data mining, and visualization tools to support informed decision making by presenting relevant data in a digestible format. 23 Predictive Analytics Characteristics – examines recent and historical data to detect patterns and predict future outcomes and trends – can forecast what might happen in the future based on probabilities – uses tools like data mining and statistical procedures such as linear regression, multiple regression, and logistic regression 24 Predictive Analytics Tools: Data Mining Data mining is used to predict trends and behaviors – In targeted marketing: data mining can use data from past promotional mailings to identify prospects that are most likely to respond favorably to future promotional mailings – In fraud detection: over time, a pattern emerges of the typical ways a client uses a credit card (where he shops, what he purchases, etc.). If the card is stolen and used fraudulently, then the usage might vary from the client’s pattern, in which case an alert is generated 25 Predictive Analytics More examples – Finance: Creating credit scores, detecting fraud, measuring credit risk, and maximizing cross-sell/up-sell opportunities – Insurance: Predicting the average costs of claims against a policy to set premiums – Healthcare: Discovering patient risk factors, determining causes of diseases, and assessing the effectiveness of drugs and techniques – Human resources: Identifying future workforce skill requirements, identifying factors that contribute to high staff turnover, and analyzing an employee's performance, skills, and preferences – Retail: Analyzing the effectiveness of promotional events and determining which offers are most appropriate for consumers – Weather: Forecasting weather – Stockbroking: Maximizing trading returns 26 Analytics Use Case 27 Prescriptive Analytics Characteristics – goes beyond descriptive and predictive analytics by recommending one or more courses of action and by identifying the likely outcome of each decision – tries to quantify the effect of future decisions – advises on the possible outcomes of a decision before it is made E.g., a driverless car must analyze the impact of a possible decision (braking, speeding, going left, etc.) before it actually makes that decision – relies on statistical procedures such as optimization, simulation, and decision trees (which are beyond the scope of this course) 28 Descriptive vs Predictive vs Prescriptive 29 Presentation tools After data is processed, it can be presented to users (decision makers) in visual formats such as text, graphics, and tables – This process is known as data visualization 30 Presentation Tools Dashboards Capability Description Drill down The ability to go to details, at several levels; it can be done by a series of menus or by clicking on a drillable portion of the screen. Critical success factors (CSFs) The factors most critical for the success of business. These can be organizational, industry, departmental, or for individual workers. Key performance indicators The specific measures of CSFs. (KPIs) Status access The latest data available on a KPI or some other metric, often in real time. Trend analysis Short-, medium-, and long-term trends of KPIs or metrics, which are projected using forecasting methods. Exception reporting Reports highlight deviations larger than defined thresholds. Reports may include only deviations. 31 Presentation Tools Dashboards – Example of performance scorecard 32 Presentation Tools GIS A Geographic Information Systems (GIS) is a computer-based system for capturing, integrating, manipulating, and displaying data using digitized maps Examples – ESRI offers GIS applications www.esri.com – Walgreens example https://www.youtube.com/watch?v=l9oXKbZQD04 – Submarine cables example https://www.youtube.com/watch?v=6dkiqJ_IZGw 33