Data is shaping culture and society.
You may not be aware of it, but business intelligence has exerted a revolutionizing influence across industries.
Data analytics was once a force in the shadows. Today insights are shaping the headlines, helping corporations, non-profits, and even Major League Baseball teams improve their organization’s performance.
1. Capitalizing Ticket Sales on Broadway
Producers from Disney Theatrical Productions are utilizing a process called dynamic ticket pricing to judge when box-office prices can be raised to extract the greatest profit without compromising attendance.
The algorithm, in use since 2011, has helped producers determine the maximum price for each of the Minskoff Theatre’s 1,700 seats.
Producers found times of year and times of day when higher prices would still sell. Disney data scientists suggest prices for five different types of shows: holiday times, such as around Christmas, off-season periods, and the times that fall between.
The analysis led to the Disney production of “The Lion King” concluding 2013 as Broadway’s top revenue-generator.
Disney uses a similarly sophisticated algorithm for other operations. Their airline and hotel combination deals are formulated for peak weeks using related software.
2. Predicting and Preventing Crime in Philadelphia
Police in Philadelphia, PA are now training and graduating their own data scientists to help them fight and prevent crime.
Using data analysis and predictive models, police have been able to more effectively prevent crimes and capture fugitives.
Now, instead of hiring data scientists to help the department keep the public safe, they are training their own officer-data experts.
The new class of experts are already police officers when they begin advanced analytics training. Department officials say homegrown experts approach data with a more intimate understanding of law enforcement.
Such interdisciplinary training could reflect a growing trend as business intelligence is assimilated by the economy.
From 2012-2013, analytics helped the program identify at-risk areas for crime and sending more regular patrols to those places. The result was a 39% reduction in home burglaries.
Such statistically-supported estimates of when and where crimes will take place
allows the department to evolve into a more preventative role.
3. Powering and Informing “Smart” Homes
Data is now teaching homes to know when residents are out of the house and when they are likely to return.
Such data-powered “smart homes” are run by climate control devices that are connected to the Internet. They sense the location of residents by geotracking their phones. This allows them to recognize when they have left the “geofence” that surrounds the property.
When all residents are gone, the heat or the air is shut off, only to be turned back to the appropriate level before they return.
As time wears on, the appliances better learn the habits of household members, and can predict with a greater degree of accuracy when they will return.
The apps can also be controlled remotely using a smartphone.
4. Turning Data into Three-Dimensional Wildlife Corridors
Tracking wildlife corridors is a notoriously tricky and inexact art for conservationists and wildlife biologists. But the art has been refined to a science detailed in 3D by predictive modeling techniques.
Informed by massive data sets gathered from GPS tracking devices attached to animals, the San Diego Supercomputer Center (SDSC) has been able to assist a conservation project working to protect the endangered California condor.
The U.S. Geologic Survey and the San Diego Institute for Conservation Research are partnered together on the project. Although they have long had access to models of the condor’s general habitat, the new models formulated with the help of the SDSC are three dimensional.
This provides greater insight on patterns of movement, and helps researchers better understand which mountainous areas are being inhabited so that they can better target their conservation.
The condor is an infamous success story, having grown from 22 birds in the 1980’s to over 400 birds today.
To view multimedia of the project, see the video posted on SeedMe.
5. Using Statistical Analysis to Drive Player Acquisition in Major League Baseball
Many professional baseball teams that are still competing for a spot in the postseason approach the trade deadline looking for a player that fits a rough profile: performing well at a position where they have a need.
The New York Yankees took that a step further, using detailed algorithms to determine the precise player that met their requirements.
That turned out to be San Diego Padres third-baseman Chase Headley. Though typical measurements did not suggest he was performing well leading up to the trade, one key statistic leapt out to General Manager Brian Cashman: velocity.
Headley has been hitting the ball harder and harder as the year went on, and Cashman leapt at the opportunity to acquire him for a relatively low price.
Cashman said that such statistics dove beneath the surface and were more reliable indicators of future performance.
Data showed that McCarthy was far more effective when using his cutter pitch. He had only used his cutter 10% of the time before the trade. Afterwards, with increased use of his cutter, he excelled in his first several starts with the Yankees.
Employing insights from big data in baseball is becoming more widespread, but it is not new. Oakland Athletics General Manager Billy Beane, who used a similar approach, was portrayed in an Oscar-nominated film about his tactics in 2011, called Moneyball.
6. Anticipating Patient Needs
University of Pittsburgh Medical Center (UPMC), a provider and insurer which operates 22 hospitals and 400 outpatient sites, deals with a massive amount of data.
Dr. Pamela Peele, Chief Analytics Officer at UPMC Health Plan, Insurance Services Division, said there was so much that it could “cripple” the organization.
Implementing business intelligence tools allowed UPMC to organize that data and transform a liability into an advantage.
Two of their member organizations were able to use the software to predict the impact of flu season on patients. They’ve also been able to figure out which patients will require the most services from their facilities, which ones are likely to require emergency care, and forecast readmission rates.
Predicting patient needs not only led to more efficient care and better service, but helped UPMC hospitals save money from regulatory fines by reducing the number of readmissions.
7. Revolutionizing Literary Analysis
Even the most profoundly traditional pastime, one rooted in manual labor and seemingly unaffected by technological growth, has been taken on by data analysis.
Some scholars of literature are employing business intelligence tools to plumb the big data created by hundreds of years of fiction and nonfiction.
The insights have been startling and controversial. They have also been insights that humans would have been reluctant to search for in traditional scholarship.
One Harvard study of five million books printed in English demonstrated that less than half the number of words that appeared in those books could be found in dictionaries. The remainder were described as “lexical dark matter.”
Another study revealed that American English, as a literary language, has become more emotional than British English.
Researchers have also used algorithms to categorize types of writers and to find outliers. For example, it was found that George Eliot, the female author of Middlemarch, wrote in patterns typically ascribed to male writers.
Reflecting the seriousness with which many scholars are regarding these new developments, the University of Nebraska has created a Center for the Digital Humanities.