Prescriptive Analytics Overview

Descriptive vs. Predictive vs. Prescriptive Analytics

• Prescriptive Analytics: Emphasizes finding the best course of action in a scenario using data. It leverages insights from descriptive and predictive analytics.
• Prescriptive Analytics Methods:
  • Graph analysis
  • Simulation
  • Complex event processing (combining data from multiple sources to model complex situations)
  • Neural networks
  • Recommendation engines
  • Heuristics
  • Machine learning
• Key Difference Between Predictive and Prescriptive Analytics: Predictive analysis helps to understand what might happen, while prescriptive analytics uses algorithms to generate automated recommendations or solutions.
• Prescriptive Analysis Considerations:
  • Prescriptive analytics requires complex algorithms for machine-based decision making.
  • There's an element of risk with automated recommendations, as human behavior can be unpredictable.

How Prescriptive Analytics Works

• Prescriptive analytics begins by identifying a problem requiring a solution.
• It then uses data to find the optimal course of action.

Prescriptive Analytics Examples in Online Learning

• Adaptive Learning: Tools that identify learned content and present new material.
• LMS (Learning Management Systems): Define rules for automated feedback or actions (e.g., recommending additional resources for struggling learners).
• Personalized Learning Paths: LMSs can use prescriptive analytics to determine learners' prior knowledge and proficiency, recommending appropriate courses or resources.

Advantages of Prescriptive Analytics

• Revenue Generation: Prescriptive analytics can identify customer preferences and new opportunities for cross-selling and sales acceleration.
• Gross Margin Management: Using prescriptive analytics alongside predictive analytics can provide insights into optimal product mixes for different market conditions.
• Expense Reduction: Techniques help manage inventory levels and reduce storage time.

Prescriptive Analysis Applications in Business

• Optimum Sourcing: Prescriptive analytics helps companies determine the best sourcing locations, logistics routes, and quantities to reduce expenses and save costs.
• Inventory Management: Considers demand and supply in the market to hold the right inventory levels, optimizing capital utilization.
• Other Examples: Self-driving cars rely on prescriptive analytics to make millions of calculations.

Why Prescriptive Analysis Matters to Businesses

• Predictive Abilities:
  • Determine the number of replacements needed based on past product sales.
  • Understand customer product preferences to launch targeted marketing campaigns.
  • Predict equipment failures, enabling proactive maintenance.
  • Analyze customer purchasing habits and payment punctuality to assess creditworthiness.

Prescriptive Analysis Benefits

• Process Optimization: Improves processes, campaigns, and strategies.
• Maintenance Management: Minimizes maintenance needs, optimizing equipment performance.
• Cost Reduction: Achieves cost reductions without impacting performance.
• Growth Planning: Increases the likelihood of companies effectively planning for internal growth.
• Other Benefits:
  • Qualitative research method analysis
  • Production optimization
  • Efficient supply chain management
  • Enhanced customer service and experience

Prescriptive Analytics Tools

• Improvado:
  • Data analytics tool designed for marketers.
  • Provides real-time data through automated dashboards and reports.
  • Customization options available.
  • Offers 150+ integrations, with custom integration options.
• RapidMiner:
  • Offers artificial intelligence and prescriptive analytics capabilities.
  • Features an open and extensive data analytics platform.
  • Includes a graphical interface for creating and maintaining predictive analytics models.
  • Supports scripting in multiple programming languages.
  • Pros:
    • Visual connection of boxes for data analysis
    • Wide range of data analytics and visualization tools
    • No coding skills required
    • Free version available
  • Cons:
    • Can be buggy at times
    • Limitations in some versions.
• Sisense:
  • Helps transform data into interactive reports.
  • Features a drag-and-drop interface for creating charts and visualizations.
  • Offers over 100 data connectors.
• Knime:
  • Open-source business intelligence (BI) tool.
  • Provides data integration, reporting, and analytics capabilities.
  • Includes a visual interface with nodes for various tasks.
  • Focuses on statistical models.
  • Pros:
    • More than 100 modules
    • Visual user interface, no coding required
    • Drag-and-drop interface for connecting nodes
  • Cons:
    • Nodes less customizable compared to Python/R libraries, though a Python/R node is available.
    • Can run slowly, especially with more extensions and nodes installed.
    • Steeper learning curve for users unfamiliar with similar tools.
• Tableau:
  • Business intelligence tool for turning data into actionable insights.
  • Drag-and-drop interface for creating interactive dashboards with visual analytics.
  • Offers two pricing options: on-premise installation and Tableau-hosted software.
  • Extensive data source connectivity.