Decision Making in Business Analytics

Chapter 1: Introduction to Decision Making in Business Analytics

Decision Making

  • Types of Decisions

    • Strategic Decisions:

    • Definition: Long-term, high-impact decisions that set the direction of the organization.

    • Time Horizon: Typically 3–5 years.

    • Example: Entering a new market.

    • Tactical Decisions:

    • Definition: Medium-term decisions that implement the goals and objectives set by its strategies.

    • Responsibility: Usually the domain of mid-level managers.

    • Example: Marketing campaigns, pricing strategies.

    • Operational Decisions:

    • Definition: Short-term, day-to-day decisions that manage daily operations.

    • Responsibility: Handled by operations managers who are closest to the customers.

    • Example: Scheduling staff, ordering supplies.

Decision-Making Process

  • The decision-making process involves 5 steps:

    1. Identify and Define the Problem: Clearly understand the issue at hand.

    2. Determine the Criteria for Evaluation: Establish the criteria used to evaluate alternative solutions.

    3. Determine Alternative Solutions: Identify various options that could be pursued.

    4. Evaluate the Alternatives: Assess the pros and cons of each option.

    5. Choose an Alternative: Make a decision based on the evaluation.

  • Managerial Decision-Making: Often influenced by tradition, intuition, and rules of thumb based on experience. However, leveraging large data sets to inform decisions can enhance effectiveness and efficiency.

Business Analytics Defined

  • Business Analytics:

    • Definition: The scientific process of transforming data into insights to improve decisions and enhance business performance.

    • Data-Driven Decision Making: Seen as more objective than other decision-making methods.

    • Tools of Business Analytics:

    • Create insights from data.

    • Improve forecasting accuracy for planning.

    • Quantify risks.

    • Yield better alternatives via analysis and optimization.

Categories of Analytical Methods and Models

  1. Descriptive Analytics:

    • Definition: Techniques that examine historical data to understand what has happened in the past.

    • Question Addressed: What happened?

  2. Predictive Analytics:

    • Definition: Methods that use models constructed from past data to predict the future or ascertain impacts of variables on each other.

    • Question Addressed: What might happen?

  3. Prescriptive Analytics:

    • Definition: Recommends actions (a decision) based on analysis to achieve desired outcomes.

    • Question Addressed: What should we do?

    • Note: Prescriptive models that rely on rules are termed rule-based models.

    • Related Concepts:

      • Competitive Advantage

      • Optimization

      • Decision Analysis

      • Simulation

      • Predictive Modeling

      • Forecasting

      • Data Mining

      • Descriptive Statistics

      • Data Visualization

      • Data Query

      • Standard Reporting

    • Degree of Complexity:

      • Categories from simplest (Descriptive) to complex (Prescriptive).

Big Data

  • Big Data:

    • Definition: Data that is too large or complex for standard data processing techniques or typical desktop software.

  • Challenges of Big Data:

    • Data storage and processing

    • Security

    • Availability of analytical talent

  • Technologies Developed:

    • Hadoop: Open-source programming environment supporting big data processing through distributed storage and cloud computing.

    • MapReduce: Programming model used in Hadoop, consisting of two major steps:

    1. Map Step: Divides tasks to be processed.

    2. Reduce Step: Aggregates results from the map phase.

4 Vs of Big Data

  • Volume:

    • Significance of electronically collected data leading to substantial quantity, e.g., companies storing over 100 terabytes of data (1 terabyte = 1,024 gigabytes).

  • Velocity:

    • The rate of data generation, collection, and processing, exemplified by real-time transactions at the New York Stock Exchange.

  • Variety:

    • Different types and sources of data generated, including quantitative, text, audio, and video.

  • Veracity:

    • Quality, accuracy, and trustworthiness of data. Questions regarding uncertainty, completeness, and correctness must be addressed.

Key Concepts in Business Analytics

  • Cloud Computing:

    • Definition: Utilization of data and software on externally housed servers via the internet, enabling feasible and cost-effective data storage and processing.

  • Data Security:

    • Definition: Protection of stored data from destructive forces or unauthorized users. Essential for confidential data, e.g., medical or financial records.

  • Artificial Intelligence (AI):

    • Application of big data and computing power to make decisions traditionally requiring human intelligence.

Ethical Issues in Analytics

  1. Privacy:

    • Necessity of protecting personal information from unauthorized access. Requirement of obtaining informed consent from data subjects about data collection and usage.

  2. Security:

    • Protecting data against breaches and unauthorized access. Compliance with laws such as GDPR (General Data Protection Regulation) and HIPAA (Health Insurance Portability and Accountability Act).

  3. Transparency:

    • Ensuring models are understandable to stakeholders, fostering trust and accountability. Clear disclosure of data sources is important.

Legal Issues in Analytics

  • Purpose Limitation:

    • Using data solely for the purposes for which it was collected.

  • Accountability:

    • Assigning clear roles for data governance and ethical practices.

  • Audits and Assessments:

    • Regular checks to maintain compliance with ethical and legal standards.

  • Ownership of Data:

    • Clearly defining who owns the data and the parameters for its use or sharing.

  • Intellectual Property Rights:

    • Respecting rights associated with data and analytics tools.

  • Legal Compliance:

    • Adherence to relevant laws governing data use and analytics, with clear agreements for data sharing with third parties.

Here are the definitions for the terms you asked about, based on the note:

Ethical Issues in Analytics

  1. Privacy: This involves protecting personal information from unauthorized access and requires obtaining informed consent from data subjects about data collection and usage.

  2. Security: This is about protecting data against breaches and unauthorized access, and it includes complying with laws such as GDPR and HIPAA.

  3. Transparency: This means ensuring that models are understandable to stakeholders, which fosters trust and accountability. Clear disclosure of data sources is also important.

Legal Issues in Analytics

  1. Purpose Limitation: This refers to using data solely for the purposes for which it was collected.

  2. Accountability: This involves assigning clear roles for data governance and ethical practices.

  3. Audits and Assessments: These are regular checks conducted to maintain compliance with ethical and legal standards.

  4. Ownership of Data: This entails clearly defining who owns the data and the parameters for its use or sharing.

  5. Intellectual Property Rights: This requires respecting rights associated