Prescriptive analytics: Difference between revisions

Revision as of 05:56, 24 August 2023

Revision as of 16:53, 13 November 2023

Line 4:

==Overview==

Prescriptive analytics is the third and final phase of business analytics, which also includes descriptive and predictive analytics.<ref>{{cite journal|author1=Evans, James R.|author2=Lindner, Carl H. |name-list-style=amp |title=Business Analytics: The Next Frontier for Decision Sciences|journal=Decision Line|date=March 2012|volume=43|issue=2}}</ref><ref name=":0">{{Cite journal |lastlast1=Basu |firstfirst1=Atanu |last2=Brown |first2=Scott |last3=Worth |first3=Tim |date=2019-10-25 |title=Predictive analytics in field service |url=http://pubsonline.informs.org/do/10.1287/LYTX.2010.06.03/full/ |journal=The Analytics Journey |language=en |doi=10.1287/lytx.2010.06.03|s2cid=242347282 }}</ref> Referred to as the "final frontier of analytic capabilities",<ref>{{Cite web |url=https://www.globys.com/2013/06/gartner-terms-prescriptive-analytics-%E2%80%9Cfinal-frontier%E2%80%9D-analytic-capabilities |title=Gartner terms Prescriptive Analytics as the "Final Frontier" of Analytic Capabilities &#124; Globys.com |access-date=2014-10-29 |archive-url=https://web.archive.org/web/20160402140918/http://globys.com/2013/06/gartner-terms-prescriptive-analytics-%E2%80%9Cfinal-frontier%E2%80%9D-analytic-capabilities |archive-date=2016-04-02 |url-status=dead }}</ref> prescriptive analytics entails the application of [[mathematical sciences|mathematical]] and [[computational science]]s and suggests decision options for how to take advantage of the results of descriptive and predictive phases.

The first stage of business analytics is descriptive analytics, which still accounts for the majority of all business analytics today.<ref>{{cite journal|last=Davenport, Tom |title=The three '..tives' of business analytics; predictive, prescriptive and descriptive|journal=CIO Enterprise Forum|date=November 2012}}</ref> Descriptive analytics looks at past performance and understands that performance by mining historical data to look for the reasons behind past success or failure. Most management reporting – such as [[sales]], [[marketing]], [[Business operations|operations]], and [[finance]] – uses this type of post-mortem analysis.

Line 13:

The final phase is prescriptive analytics,<ref>{{cite journal|last1=Haas|first1=Peter J.|author1-link=Peter J. Haas (computer scientist)|last2=Maglio|first2=Paul P.|last3=Selinger|first3=Patricia G.|author3-link=Patricia Selinger|last4=Tan|first4=Wang-Chie|issue=12|journal=Proceedings of the VLDB Endowment|title=Data is Dead…Without What-If Models|volume=4|year=2011|pages=1486–1489|doi=10.14778/3402755.3402802|s2cid=6239043|doi-access=free}}</ref> which goes beyond predicting future outcomes but also suggesting actions to benefit from the predictions and showing the implications of each decision option.<ref>{{cite journal |author1=Stewart, Thomas. R. |author2=McMillan, Claude, Jr. |name-list-style=amp |year=1987 |title=Descriptive and Prescriptive Models for Judgment and Decision Making: Implications for Knowledge Engineering |journal=Expert Judgment and Expert Systems |volume=NATO AS1 Subseries F35 |pages=314–318}}</ref>

Prescriptive analytics uses algorithms and machine learning models to simulate various scenarios and predict the likely outcomes of different decisions.<ref name=":1">{{Citation |lastlast1=Soltanpoor |firstfirst1=Reza |title=Prescriptive Analytics for Big Data |date=2016 |url=http://link.springer.com/10.1007/978-3-319-46922-5_19 |work=Databases Theory and Applications |volume=9877 |pages=245–256 |editor-last=Cheema |editor-first=Muhammad Aamir |access-date=2023-05-01 |place=Cham |publisher=Springer International Publishing |doi=10.1007/978-3-319-46922-5_19 |isbn=978-3-319-46921-8 |last2=Sellis |first2=Timos |series=Lecture Notes in Computer Science |editor2-last=Zhang |editor2-first=Wenjie |editor3-last=Chang |editor3-first=Lijun}}</ref> It then suggests the best course of action based on the desired outcome and the constraints of the situation. Prescriptive analytics not only anticipates what will happen and when it will happen, but also why it will happen.<ref name=":1" /> Further, prescriptive analytics suggests decision options on how to take advantage of a future opportunity or mitigate a future risk and shows the implication of each decision option. Prescriptive analytics incorporates both [[structured data|structured]] and [[unstructured data]], and uses a combination of advanced analytic techniques and disciplines to predict, prescribe, and adapt. It can continually take in new data to re-predict and re-prescribe, thus automatically improving prediction accuracy<!-- This lacks evidence and supporting citation. It does not follow that prediction accuracy improves as a result of re-predicting. --> and prescribing better decision options. Effective prescriptive analytics utilises hybrid data, a combination of structured (numbers, categories) and unstructured data (videos, images, sounds, texts), and business rules to predict what lies ahead and to prescribe how to take advantage of this predicted future without compromising other priorities.<ref>{{cite journal |last1=Riabacke |first1=Mona |last2=Danielson |first2=Mats |last3=Ekenberg |first3=Love |title=State-of-the-Art Prescriptive Criteria Weight Elicitation |journal=Advances in Decision Sciences |date=30 December 2012 |volume=2012 |pages=1–24 |doi=10.1155/2012/276584 |doi-access=free }}</ref> Basu suggests that without hybrid data input, the benefits of prescriptive analytics are limited.<ref name="basu" />{{efn|Atanu Basu is the CEO and president of Ayata.<ref name=basu />}}

In addition to this variety of data types and growing data volume, incoming data can also evolve with respect to velocity, that is, more data being generated at a faster or a variable pace. Business rules define the [[business process]] and include objectives constraints, 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 resources on a massive scale including human, computational and temporal for every Prescriptive Analytic project. In order to spare the expense of dozens of people, high performance machines and weeks of work one must consider the reduction of resources and therefore a reduction in the accuracy or reliability of the outcome. The preferable route is a reduction that produces a probabilistic result within acceptable limits.{{citation needed|date=May 2020}}