Prescriptive analytics

Prescriptive analytics is the third and final phase of business analytics , which also includes descriptive and predictive analytics. [1] [2]

Referred to the “final frontier of analytic capabilities,” [3] prescriptive analytics entails the application of mathematical and computational sciences and suggests decisions to take advantage of the results of descriptive and predictive analytics. The first stage of business analytics is descriptive analytics, which still accounts for the majority of all business analytics today. [4] Descriptive analytics looks at past performance and understands that past performance or mining. Most management reporting – such as sales , marketing , operations , and finance – uses this type of post-mortem analysis.

Prescriptive Analytics extends beyond predictive analytics by specifying both the actions necessary to achieve predicted outcomes, and the interrelated effects of each decision

The next phase is predictive analytics . Predictive analytics answers the question what is likely to happen. This is when the data is combined with rules, algorithms , and likely to determine the likely outcome of an event or the likelihood of a situation. The final phase is prescriptive analytics, [5] which goes beyond predicting future outcomes, and includes actions to benefit the predictions and showing the implications of each decision option. [6]

Prescriptive analytics not only anticipates what will happen and when it will happen, but also why it will happen. Further, prescriptive analytics suggests a choice of future options and options for future decision-making. Prescriptive analytics can continually take over new data to re-predict and re-prescribe, thus automatically improving prediction accuracy and prescribing better decision options. Prescriptive analytics ingests hybrid data, a combination of structured (numbers, categories) and unstructured data (videos, images, sounds, texts), and business rules to predictable future predictions and predictions Priorities. [7]

All three phases of analytics can be performed through professional services or technology or a combination. In order to scale, prescriptive analytics technologies need to be adapted to take into account the growing volume, velocity, and variety of data that most mission critical processes and their environment may produce.

One criticism of prescriptive analytics is that its distinction from predictive analytics is ill-defined and therefore ill-conceived. [8]


Prescriptive analytics incorporates both structured and unstructured data, and uses a combination of advanced analytic techniques and disciplines to predict, prescribe, and adapt. While the term is prescriptive analytics was first coined by IBM [2] and later trademarked by Ayata, [9] the underlying concepts have been around for hundreds of years. The technology behind prescriptive analytics combines hybrid data , business rules with mathematical models and computational models. The data inputs to prescriptive analytics may come from multiple sources: internal, such as inside a corporation; and external, also known as environmental data. The data may be structured, which includes numbers and categories, as well as unstructured data , such as texts, images, sounds, and videos. Unstructured data differs from structured data in its various formats and can not be stored in traditional databases. [10] More than 80% of the world’s data today is unstructured, according to IBM.

In addition to this type of data and growing data volume, the data can also evolve with respect to velocity, which is more data being generated at a faster or a variable pace. Business rules define the business processand include objectives, preferences, policies, best practices, and boundaries. Mathematical models and computational models are techniques derived from mathematical sciences, computer science and related disciplines such as applied statistics, machine learning, operations research, natural language processing, computer vision, pattern recognition, image processing, speech recognition, and signal processing. The correct application of all these methods and the verification of their results implies the need for a massive scale of human resources, computational and temporal for every Prescriptive Analytic project. In order to spare the expense of dozens of people, high performance machines and long-term care must consider the reduction of resources and therefore reduce the accuracy or reliability of the outcome.

Applications in Oil and Gas

Energy is the largest industry in the world ($ 6 trillion in size). The processes and decisions related to oil and natural gas exploration, development and production. Many types of data are being used in the field of solar energy, such as depositional characteristics, . [11] Prescriptive analytics software can help with both locating and producing hydrocarbons [12]

by taking in seismic data, well log data, production data, and other related data sets to prescribe specific recipes for how and where to drill, complete, and produce well in order to optimize recovery, minimize cost, and reduce environmental footprint. [13]

Unconventional Resource Development

Examples of structured and unstructured data sets generated by the oil and gas companies and their ecosystem services that can be analyzed using Prescriptive Analytics software

With the value of the end product determined by global commodity economics, the basis of competition for operators in the E & P market is the ability to effectively deploy capital more effectively and predictably, predictably, and safely than their peers. In unconventional resource plays, operational efficiency and effectiveness is diminished by reservoir inconsistencies, and decision-making uneven by high degrees of uncertainty. These challenges manifest themselves in the form of low recovery factors and wide performance variations.

Prescriptive Analytics software can accurately predict production and prescribe the optimal configurations of controllable drilling, completion, and production variables by variable internal and external variables, regardless of source, structure, size, or format. [14] Prescriptive analytics software can also provide decision options and show the impact of each decision option to the operations managers can proactively take appropriate actions, on time, to guarantee future exploration and production performance, and maximize the economic value of assets at every point over the course of their serviceable lifetimes. [15]

Oilfield Equipment Maintenance

In the realm of oilfield equipment maintenance, Prescriptive Analytics can optimize configuration, anticipate and prevent unplanned downtime, optimize field scheduling, and improve maintenance planning. [16] According to General Electric, there are more than 130,000 electric submersible pumps (ESPs) installed globally, accounting for 60% of the world’s oil production. [17] Prescriptive Analytics has been deployed to predict when and why. [18]

In the area of health, safety, and the environment , prescriptive analytics can predict and preempt incidents that can lead to reputational and financial loss for oil and gas companies.


Pricing is another area of ​​focus. Natural gas prices fluctuate dramatically on supply, demand, econometrics , geopolitics , and weather conditions. Gas producers, pipeline transmission companies and utility firms have a keen interest in more accurate predicting gas prices so that they can lock in favorable terms while hedging downside risk. Prescriptive analytics can be used to estimate the options and show the impact of each decision option. [19]

Applications in healthcare

Multiple factors are driving healthcare providers to Dramatically Improve business processes and operations as the United States healthcare industry embarks on the migration from Necessary Largely a fee-for service volume-based system to a fee-for-performance, value-based system. Prescriptive analytics is playing a key role to help the performance in a number of areas involving various stakeholders: payers, providers and pharmaceutical companies.

Prescription analytics can help providers improve their ability to achieve better patient satisfaction and retention. Providers can do better population health management by identifying appropriate intervention models for risk stratified population combining data from the in-facility care episodes and home based telehealth.

Prescriptive analytics can also benefit healthcare providers in their capacity planning, and the use of these data is more likely to be used in the future. A new facility with existing facilities versus a new one. [20]

They are most suitable for clinical trials worldwide – patients who are expected to be compliant and will not drop out of the trial due to complications. Analytics can tell companies how much time and money they can save if they choose a patient cohort in a specific country. Reviews another.

In provider-pay negotiations, providers can improve their negotiating position with health insurers by developing a robust understanding of future service utilization. By accurately predicting utilization, providers can also better allocate staff.

See also

  • Analytics
  • Applied Statistics
  • Big Data
  • Business analytics
  • Business Intelligence
  • Data mining
  • Decision Management
  • Decision Engineering
  • Forecasting
  • Hadoop
  • MapReduce
  • OLTP
  • Operations Research
  • Statistics


  1. Jump up^ Evans, R. James & Lindner, Carl H. (March 2012). “Business Analytics: The Next Frontier for Decision Sciences”. Decision Line . 43 (2).
  2. ^ Jump up to:b Lustig, Irv, Dietrich, Brenda Johnson, Christer, and Dziekan, Christopher (Nov. -Dec 2010). “The Analytics Journey”. Analytics .
  3. Jump up^
  4. Jump up^ Davenport, Tom (November 2012). “The three ‘..’ business analytics, predictive, prescriptive and descriptive ‘. CIO Enterprise Forum .
  5. Jump up^ Haas, Peter J., Maglio, Paul P., Selinger, Patricia G., and Tan, Wang-Chie (2011). “Data is Dead … Without What-If Models”. Proceedings of the VLDB Endowment . 4 (12).
  6. Jump up^ Stewart, Thomas. R. & McMillan, Claude, Jr. (1987). “Descriptive and Prescriptive Models for Decision Making and Decision Making: Implications for Knowledge Engineering”. NATO AS1 Senes, Expert Judgment and Expert Systems . F35 : 314-318.
  7. Jump up^ Riabacke, Mona, Danielson, Mats, and Ekenber, Love (2012). “State-of-the-Art Prescriptive Criteria Weight Elicitation”. Advances in Decision Sciences .
  8. Jump up^ Bill Vorhies (November 2014). “Prescriptive vs. Predictive Analytics – A Distinction Without a Difference?” . Predictive Analytics Times .
  9. Jump up^
  10. Jump up^ Inmon, Bill; Nesavich, Anthony (2007). Tapping Into Unstructured Data . Prentice-Hall. ISBN  978-0-13-236029-6 .
  11. Jump up^ Basu, Atanu (November 2012). “How Prescriptive Analytics Can Reshape Fracking in Oil and Gas Fields”. Data-Informed .
  12. Jump up^ Basu, Atanu (December 2013). “How Data Analytics Can Help Frackers Find Oil”. Datanami .
  13. Jump up^ Mohan, Daniel (August 2014). “Machines Prescribing Recipes from ‘Things,’ Earth, and People ‘. Oil & Gas Investor .
  14. Jump up^ Basu, Mohan, Marshall, & McColpin (December 23, 2014). “The Journey to Designer Wells”. Oil & Gas Investor .
  15. Jump up^ Mohan, Daniel (September 2014). “Your Data Already Know What You Do not”. E & P Magazine .
  16. Jump up^ Presley, Jennifer (July 1, 2013). “ESP for ESPs”. Exploration & Production.
  17. Jump up^
  18. Jump up^ Wheatley, Malcolm (May 29, 2013). “Underground Analytics”. DataInformed .
  19. Jump up^ Watson, Michael (November 2012). “Advanced Analytics in Supply Chain – What Is It Better than Non-Advanced Analytics?” Supply Chain Digest .
  20. Jump up^ Foster, Roger (May 2012). “Big data and public health, part 2: Reducing Unwarranted Services”. Government Health IT .