A beautiful future depends on data and AI

01

A festival rocks with data

Imagine the extraordinary challenge of creating an impromptu city for tens of thousands of devoted music fans on 148 acres of ancient beech forest in the Danish countryside. Sound incredible? Yet this is exactly what Smukfest — “Denmark’s most beautiful festival” — has done for the past four decades.

Founded in 1980 when five young men gathered a few bands and several hundred fans in the forest for a day of music, Skanderborg Music Festival — Smukfest — has become the second largest music festival in Denmark, hosting national favorites and international artists like Eric Clapton, Britney Spears, Shawn Mendez and deadmau5.

For six weeks every summer, as many as 14,000 dedicated festival volunteers nearly double the population of nearby Skanderborg, readying the site in the forest of Dyrehaven for the arrival of 170 acts and 60,000 jubilant Smukfest fans. A wildly popular event, tickets to the full-week festival sell out in the first hour of sales. Suffused with an atmosphere of free-spirited whimsy — famous for zany events like the Naked Canoe race — Smukfest delivers a welcome escape to peaceful — though highly energized — sylvan surrounds.

True to its idealistic origins, the festival is owned and managed by a nonprofit organization comprising many thousands of volunteers who bring this unforgettable event to life every August. The infrastructure is environmentally friendly, the festival establishment socially conscious. All festival bars, shops and installations are crafted from 100 percent recyclable materials, and 20 percent of proceeds fund a national nonprofit to benefit vulnerable people, with special programs focused on loneliness in youth.

Creating a week-long celebration of “love, life and togetherness” for 60,000 enthused concertgoers — “old and young, CEOs and students, hippies n’ yuppies and everything in between” — is no small feat.

Creating a week-long celebration of “love, life and togetherness” for 60,000 enthused concertgoers — “old and young, CEOs and students, hippies n’ yuppies and everything in between” — is no small feat.

Creating a week-long celebration of “love, life and togetherness” for 60,000 enthused concertgoers — “old and young, CEOs and students, hippies n’ yuppies and everything in between” — is no small feat. Festival organizing involves more than booking bands and staging performances. Every logistical challenge posed by the management of a small city — transportation services and traffic management, food and beverage provision, sanitation, lodging and accommodations, communications, security and emergency services, sales and supply chain — must be addressed on site, in real time.

Little did Smukfest founders imagine their modest event in the woods would become an international entertainment venue the size of a small city. As the festival grew, organizers faced increasingly more complex logistical challenges. About a decade ago, Smukfest began implementing a few electronic enhancements to improve festival operations. Web support for lodging helped organizers streamline attendee accommodations; later, the festival added near field communication (NFC) chip bracelets and bartender robots to its digital lineup.

With these and other additions over the years, the festival amassed vast volumes and variety of data about its operations, and especially about attendee behavior. Organizers recognized the key to better festival management lay in uncovering insights buried in vast troves of data generated by participants themselves. These challenges and solutions formed two sides of the same coin: Smukfest planners knew that analyzing this data could simultaneously transform operations and the fan experience.

Though a smooth-running festival requires muscle, sweat and elbow grease, the fuel that powers this engine of free-spirited togetherness is data, and plenty of it.

02

AI celebrates “love, life and togetherness”

In 2018, Smukfest joined forces with IBM® Denmark to bring greater sophistication to its data collection and analysis. A group of IBM volunteers, many longtime participants at Smukfest, left the comfort of their offices and spent eight days working in portable containers and camping in the forest alongside festivalgoers. The team’s goal was to collect and analyze festival data and iterate solutions in real time to help build a better fan experience.

“The festival is already on the leading edge,” said Lars B. Pedersen, senior consultant with IBM Global Business Services®, “collecting data about everyone in the audience, including ticketing, purchases, payments, location data, social media, music preferences and weather data.”

The festival is already on the leading edge, collecting data about everyone in the audience, including ticketing, purchases, payments, location data, social media, music preferences and weather data.”

Lars B. Pedersen
Senior consultant
IBM Global Business Services

The festival is already on the leading edge, collecting data about everyone in the audience, including ticketing, purchases, payments, location data, social media, music preferences and weather data.”

Lars B. Pedersen
Senior consultant
IBM Global Business Services

Scores of IBM volunteers bring unprecedented insight to festival operations

Even before Smukfest and IBM initiated a working relationship, scores of IBM Denmark employees had volunteered personal time at the festival over several decades. Those informal connections, with IBM employees observing and participating in festival operations, inevitably yielded a deeper understanding of the organization’s information architecture. IBM employees offered suggestions for how IBM solutions could help create a better experience for the audience, helping Smukfest deliver on its strategic vision.

03

Insights revealed, privacy kept

Using a data lake to pool information

If data is the new electricity, Smukfest possesses enough to power a small city. To harness that energy, data must be consolidated and accessible. The IBM team created a data lake to pool relevant information that could be accessed easily by different applications and for different purposes.

We had this huge amount of data in the forest, but it was in silos. We didn’t have access to all the data or any chance to combine data. We’re not just organizing and analyzing data but combining different kinds of data sets to give us different points of view that can support our decision making.”

Niels Mørk
CIO
Smukfest

We had this huge amount of data in the forest, but it was in silos. We didn’t have access to all the data or any chance to combine data. We’re not just organizing and analyzing data but combining different kinds of data sets to give us different points of view that can support our decision making.”

Niels Mørk
CIO
Smukfest

IBM teams gleaned data from attendee activities, and in compliance with EU General Data Protection Regulation (GDPR), cleansed all digital traces of personal information. With personal information removed, the IBM team could perform sophisticated analytics on the accumulated information without restrictions.

Design thinking promotes more agile organization

Using design thinking and agile methodology, the IBM team worked around the clock alongside Smukfest organizers to devise a set of use cases to study and solve key organizational challenges. The team collected and analyzed huge troves of data captured during the festival, revealing insights that could yield a better festival experience. Analyses helped reveal attendee food and beverage purchasing patterns, improve crowd safety, indicate audience preferences and address frequently asked questions.

The solutions the teams developed and implemented at Smukfest 2018 delivered a variety of benefits, helping teams envision and design use cases for exploration at the 2019 festival. To collect and analyze festival data, the team used IBM Streaming Analytics for IBM Cloud to prepare data; built databases powered by IBM Compose; captured and stored the data using IBM Cloud™ Object Storage; with IBM Watson® Studio supported data analysis, and IBM Event Streams powered festival apps.

04

The most beloved bar drink?

Analyzing data to calibrate festival operations

For the 2019 festival, Smukfest and IBM teams crafted a series of queries to probe deeply into the dizzying volume of data gleaned from current and previous festival years. For instance, are Smukfest beverage sales correlated to musical genres or scheduled performances? How do bar locations impact sales? Are beverage types associated with particular performers?

To find these answers, IBM volunteers dove into the tremendous trove of data collected at the festival. The rich volume and variety provided teams the opportunity to examine data from multiple points of view and correlate relationships across seemingly disparate categories. IBM festival volunteers collected data by time, venue, musical genre and sales. Associating these data categories to understand product sales at bars during events and combining each data point to create dashboard visualizations enabled teams to gain evidence-driven insight to guide real-time — and future — decision making.

With IBM Watson technology, existing data is combined with new enriched data sources, such as motion and weather sensors. The ability of IBM Watson to handle large data sets with many different data types from IoT devices enables the Smukfest team to make instant decisions in real time. Smukfest issues digital wristbands equipped with near field communication technology to festivalgoers to provide secure, cashless transactions for ticketing, food and beverage purchases, accommodations and VIP festival access, producing nearly 750,000 cashless transactions and a goldmine of data. And with nearly every participant carrying a smartphone, the number of sensor and data points is amplified exponentially throughout the massive crowd, providing more precise data to help the team further refine its analyses. Add to these data sources Spotify playlists, identifying real-time information about favorite groups and songs, and geolocation metadata gleaned from smartphone movement to improve deployment for security and medical teams.

Using data to predict inventory needs and diminish waste

Festival foodies can recharge with fresh oysters and champagne, tapas and sangria, gourmet burgers and craft beer at scores of pop-up eateries, 65 bars and 600 points of sale, all paid with a flick of the wristband. The flip side of cashless abundance and convenience is an astounding profusion of food waste. In 2018, nearly 40 percent of 325,000 food items were returned after the festival, costing precious time and money.

To forecast 2019 product sales, the IBM team used eight different data sources and 34 data preparation steps to transform raw and granular 2018 transaction data into a clean, aggregated data set. Data about hourly and daily sales of products at bars was compared against weather data and music schedules, revealing invaluable insights about the texture of participant behavior and interactions that help organizers plan to an astounding level of granular accuracy. Using multidimensional forecasting models incorporating information from weather to sales, IBM helped Smukfest planners predict inventory needs, reducing errors that can generate over- or understocking critical products, enabling vendors to place more precise orders for inventory based on real-time conditions. The teams envision that at future festivals, bar managers will be able to depend on data and AI to provide reliable inventory recommendations — such as how many oranges to stock to keep crowds refreshed and happy.

Data collected, organized and available

Collected, organized and put to work, Smukfest data is available for use with any application available on IBM Cloud Pak™ for Data, from machine learning capabilities inside Watson Studio to model testing, bias detection and outcome explainability in IBM Watson® OpenScale™.

Using IBM data and AI offerings and collecting and combining data in the cloud, Smukfest continues to glean unique insights in real time about a comprehensive range of attendee behaviors and interactions at the festival. With the festival’s cooperation, the solutions created at the festival will yield innovations applicable to other industries including entertainment, retail, consumer products and banking, with crowd safety data relevant for law enforcement and defense.