Predictive, descriptive and

prescriptive models:
What they are and how to
apply them in business
Predictive, Descriptive and Prescriptive
Models
• Predictive, Descriptive and Prescriptive models are key
elements in data science and artificial intelligence.
• They are analytical models that provide a comprehensive
framework for predicting, understanding and making data-
driven decisions.
• AI, today, offers many opportunities for improving the
efficiency and effectiveness of business operations, optimizing
decision-making processes and industrial automations, and
enabling companies to remain competitive in the marketplace,
across industries.
Predictive models
• Predictive models are designed to make predictions or
estimates about future events.
• Using advanced machine learning algorithms, these
models analyze historical data to identify patterns and
trends, and provide intuitive data-driven information and
estimates about the likelihood of a future outcome.
• For example, a predictive model could be used to estimate
future sales of a product based on historical sales data.
• Predictive analyses are then used when one wants
to understand the future.
Descriptive models
• Descriptive models are designed to describe and understand
ongoing phenomena.
• These models use statistical analysis techniques to extract
meaningful information from the data, such as mean, standard
deviation, distribution, and correlations.
• Descriptive models provide an overview of the past and present,
helping to identify patterns and trends in data behavior.
• They are often used for exploratory analysis and to gain an in-depth
understanding of the characteristics of the data themselves; in this
sense, they provide a knowledge base on which to build nowcasting
models.
• Descriptive analyses are then used when one wants to understand,
at an aggregate level, what is happening in an organization.
Prescriptive models
• Prescriptive models, on the other hand, are designed to provide
recommendations or guidance on what to do based on the
available data and drive activity toward a solution.
• These models use complex algorithms to process data
and generate suggestions about possible actions to take,
attempting to quantify the effect of future decisions to
recommend possible outcomes before they are actually taken.
• For example, a prescriptive model can be used to suggest the
best pricing strategy to maximize a company’s profits,
considering various factors such as cost, demand and
competition.
• Thus, prescriptive analyses are used when one wants
to provide recommendations regarding actions.
Applications of mathematical models in
management and industrial processes
• Predictive, descriptive, and prescriptive mathematical models
have various practical applications both in the management of
“office” procedures – such as HR, document area, budget
allocation etc. – as well as in industrial and manufacturing
processes.
Here are some real-world applications
examples:
• Demand forecasting – a company can use predictive models to
estimate future demand for a product or service. This can help in
optimizing production strategies, inventory planning and supply
chain management;
• Market data analysis – descriptive models can be used to
analyze market data and understand customer behavior, buying
preferences, and market trends. This information can be
leveraged to adapt marketing strategies and product positioning;
• Manufacturing process efficiency – prescriptive models can be
used to optimize industrial production processes. For example, a
prescriptive model can suggest the most efficient processing
sequence, minimizing production time and reducing waste;
Here are some real-world applications
examples:
• Maintenance management – predictive models can be used to
predict failures or malfunctions in industrial equipment. This
allows preventive maintenance to be scheduled to minimize
unplanned shutdowns and optimize operational efficiency;
• Rationalization of human resources – prescriptive models can be
applied to HR management, for example, to optimize staff
scheduling, resource allocation, and prediction of skills needed for
future projects;
• Optimization of workflow and business decisions – prescriptive
models can help to make optimal decisions regarding resource
allocation, activity planning and monitoring of business KPIs. For
example, they can be used to determine pricing strategies,
procurement strategies and risk management policies.
Comparing Descriptive, Predictive, and Prescriptive
Analytics Models
• Data analytics is the process of examining datasets to draw
conclusions about the information they contain.
• It involves various techniques and tools to analyze raw data
and extract meaningful insights.
• The primary goal of data analytics is to support decision-
making by providing actionable insights.
• The three main types of data analytics models are descriptive,
predictive, and prescriptive analytics each serving a unique
purpose and providing different insights.
More on What is Descriptive Analytics?
• Descriptive analytics is the process of analyzing historical data
to understand what has happened in the past.
• It focuses on summarizing and interpreting data to provide
insights into past performance and trends.
• Descriptive analytics answers the question, “What
happened?”
Techniques and Tools for Descriptive Analytics
• Descriptive analytics employs various techniques and tools,
including:
• Data Aggregation: Combining data from multiple sources to
provide a comprehensive view.
• Data Mining: Extracting patterns and relationships from large
datasets.
• Data Visualization: Using charts, graphs, and dashboards to
represent data visually.
• Statistical Analysis: Applying statistical methods to summarize and
describe data.
Common tools used in descriptive analytics
include:
• Excel: For basic data analysis and visualization.
• Tableau: For advanced data visualization and dashboard
creation.
• Power BI: For interactive data visualization and business
intelligence.
• SQL: For querying and managing databases.
Applications of Descriptive Analytics
• Descriptive analytics is widely used across various industries
for:
• Business Reporting: Generating regular reports on sales,
revenue, and other key performance indicators (KPIs).
• Customer Segmentation: Analyzing customer data to identify
different segments and their characteristics.
• Market Analysis: Understanding market trends and consumer
behavior.
• Operational Efficiency: Monitoring and improving business
processes.
What is Predictive Analytics?
• Predictive analytics uses historical data and statistical
algorithms to forecast future events. It aims to predict what is
likely to happen based on past trends and patterns.
• Predictive analytics answers the question, “What could
happen?”
Techniques and Tools for Predictive Analytics
• Predictive analytics involves several techniques and tools,
including:
• Regression Analysis: Modeling the relationship between
dependent and independent variables.
• Time Series Analysis: Analyzing data points collected or recorded
at specific time intervals.
• Machine Learning: Using algorithms to learn from data and make
predictions.
• Classification and Clustering: Grouping data into categories or
clusters based on similarities.
Common tools used in predictive analytics
include:
• R: For statistical computing and graphics.
• Python: For machine learning and data analysis libraries like scikit-
learn and TensorFlow.
• SAS: For advanced analytics, business intelligence, and data
management.
• IBM SPSS: For statistical analysis and predictive modeling.
Applications of Predictive Analytics
• Predictive analytics is applied in various fields, such as:
• Risk Management: Predicting potential risks and their impact on
business operations.
• Customer Retention: Identifying customers at risk of churning and
developing retention strategies.
• Sales Forecasting: Estimating future sales based on historical
data.
• Healthcare: Predicting disease outbreaks and patient outcomes.
What is Prescriptive Analytics?
• Prescriptive analytics goes beyond predicting future outcomes
by recommending actions to achieve desired results.
• It combines data, algorithms, and business rules to suggest the
best course of action.
• Prescriptive analytics answers the question, “What should we
do?”
Techniques and Tools for Prescriptive Analytics
• Prescriptive analytics utilizes various techniques and tools, including:
• Optimization: Finding the best solution from a set of feasible
options.
• Simulation: Modeling complex systems to evaluate different
scenarios.
• Decision Analysis: Assessing and comparing different decision
options.
• Machine Learning: Using algorithms to learn from data and make
recommendations.
Common tools used in prescriptive
analytics include:
• Gurobi: For mathematical optimization.
• IBM ILOG CPLEX: For optimization and decision support.
• AnyLogic: For simulation modeling.
• MATLAB: For numerical computing and optimization.
Applications of Prescriptive Analytics
• Prescriptive analytics is used in various industries for:
• Supply Chain Optimization: Improving inventory management
and logistics.
• Revenue Management: Setting optimal pricing strategies.
• Healthcare: Recommending personalized treatment plans.
• Finance: Optimizing investment portfolios and risk
management strategies.
Key Differences Between Descriptive, Predictive
and Prescriptive data analytics model
• While descriptive, predictive, and prescriptive analytics are
interconnected, they serve different purposes and provide different
insights:
• Descriptive Analytics: Focuses on understanding past events and
trends. It provides a summary of historical data and helps identify
patterns and relationships.
• Predictive Analytics: Uses historical data to forecast future events. It
helps anticipate potential outcomes and trends, enabling proactive
decision-making.
• Prescriptive Analytics: Recommends actions to achieve desired
outcomes. It combines data, algorithms, and business rules to suggest
the best course of action.
 The key differences can be summarized as follows:
Descriptive Predictive Prescriptive

Predictive data analytical model Prescriptive data analytical

Descriptive data analytical
use statistical models and model use optimization and
model use data aggregation and
forecast techniques to simulation algorithms to advice
data mining to provide insight
understand the future and on possible outcomes and
into the past and answer.
answer. answer.

It focuses on – What has It focuses on – What could It focuses on – What should we

happened in the past? happen in the future? do?

This is the analysis of the past or This analysis showcases feasible

historical data used to This analysis is used to solutions to a problem and the
understand the trends and determine the future trends. impact of considering a solution
estimate metrics over time. on the future trend.
 The key differences can be summarized as follows:
Descriptive Predictive Prescriptive

It is used when the user It is used when the user

It is used when the user
want to summarize results want to make an educated
have to make complex or
for all part or a part of guess at the likely
time sensitive decisions.
business. outcomes.

Tools Used – data mining, Tools Used – machine Tools Used – heuristics,
data aggregation learning, statistical model optimization

Use reactive approach. Use proactive approach. Use proactive approach.

The key differences can be summarized as follows:

Descriptive Predictive Prescriptive

Example:
Ecommerce Example:
Example : businesses which uses Identifying techniques
Annual Revenue Report the customer’s to optimize the patient
browsing history to care in the healthcare
recommend products
Conclusion
• Descriptive, Predictive and Prescriptive data analytics are
important types of analytics where Descriptive analytics is
used to summarize the data, Predictive analytics is used to
make future predictions based on the past data and
Prescriptive analytics is used to identify the possible future
outcomes and show the best option.