What is Prescriptive Analytics
Prescriptive analytics is the use of technology to help businesses make better decisions about how to handle specific situations by factoring in knowledge of possible situations, available resources, past performance and what is currently happening. Prescriptive analytics works with predictive analytics — the use of statistics and modeling to determine future performance based on current and historical data — to improve business decisions despite uncertainty and changing conditions, and to help companies determine what action to take. Prescriptive analytics can help prevent fraud, limit risk, increase efficiency, meet business goals and create more loyal customers. It can be used to make decisions on any time horizon, from immediate to long term.
It offers a vast improvement over descriptive analytics, which examines decisions and outcomes after the fact.
BREAKING DOWN Prescriptive Analytics
Prescriptive analytics relies on artificial intelligence techniques, such as machine learning — the ability of a computer program, without additional human input — to learn from and adapt to new data. Machine learning makes it possible to process the tremendous amount of data available today so that companies can learn from it. As new or additional data becomes available, computer programs adjust automatically to make use of it. This process far exceeds human capabilities because it is much faster and more comprehensive. Prescriptive analytics makes use of machine learning to help businesses decide what the importance of a new development is and how to react to it based on a computer program’s predictions.
How Prescriptive Analytics Works
Numerous types of data-intensive businesses can benefit from using prescriptive analytics, including financial services, healthcare and government. It is especially valuable in fields like these because of the high cost of human errors. Prescriptive analytics is not foolproof, however. It is only effective if organizations know what questions to ask and how to react to the answers. If the input assumptions are invalid, the output results will not be accurate.
When used effectively, however, prescriptive analytics can help organizations make decisions based on highly analyzed facts rather than jump to under-informed conclusions based on instinct. Prescriptive analytics can simulate the probability of various outcomes and show the probability of each, helping organizations to better understand the level of risk and uncertainty they face than they could be relying on averages. Organizations can gain better understandings of the likelihood of worst-case scenarios and plan accordingly.
Examples of Prescriptive Analytics
Prescriptive analytics could be used to evaluate whether a local fire department should require residents to evacuate a particular area when a wildfire is burning nearby. It could also be used to predict whether an article on a particular topic will be popular with readers based on data about searches and social shares for related topics. Another use could be to adjust a worker training program in real time based on how the worker is responding to each lesson.
Similarly, prescriptive analytics can be used to improve patient outcomes in healthcare. It puts healthcare data in context to evaluate the cost effectiveness of procedures and treatments and to evaluate official clinical procedures. It can also be used to analyze which hospital patients have the highest risk of re-admission so that healthcare providers can do more to stave off re-admission through patient education and doctor follow-up.
Suppose you are the CEO of an airline and you want to maximize your company’s profits. Prescriptive analytics can help you do this by automatically adjusting ticket price and availability based on numerous factors, including customer demand, weather, and oil prices. When the algorithm identifies that this year’s pre-Christmas ticket sales from Los Angeles to New York are lagging last year’s, for example, it can automatically lower prices, while making sure not to drop them too low in light of this year’s higher oil prices. At the same time, when the algorithm evaluates the higher than usual demand for tickets from St. Louis to Chicago because of icy road conditions, it can raise ticket prices automatically. The CEO doesn’t have to stare at a computer all day looking at what’s happening with ticket sales and market conditions and then instruct workers to log into the system and change the prices manually; a computer program can do all of this and more at a faster pace.