Dataiku Case Studies Anomaly Detection: How to Improve Core Product Accuracy and Efficiency with IoT

Edit This Case Study Record

	Anomaly Detection: How to Improve Core Product Accuracy and Efficiency with IoT Dataiku

Anomaly Detection: How to Improve Core Product Accuracy and Efficiency with IoT

Dataiku

Technology Category	Analytics & Modeling - Predictive Analytics Analytics & Modeling - Real Time Analytics
Applicable Industries	Transportation
Applicable Functions	Logistics & Transportation
Use Cases	Real-Time Location System (RTLS) Predictive Maintenance
Services	Data Science Services
Challenge	Coyote, the European leader of real-time road information, faced a significant challenge in maintaining the accuracy of speed limit data within their embedded maps. This data is crucial for the functioning of their IoT devices and mobile applications, which warn drivers of traffic hazards and conditions. The company needed an automated, algorithm-based solution that could correct speed limit data and leverage the high volume of incoming data from their IoT devices to generate actionable insights and predictions. This also required instilling a data-driven approach within the company, where decisions are based on real-world data rather than standard analytics reports. Read More
About Customer	Coyote is the European leader in real-time road information. The company was founded in 2005 and has 250 employees. It boasts 4.8 million users across Europe and generated a turnover of over €100 million in 2014. Coyote's products include IoT-based devices and mobile applications that enable users to warn other drivers of traffic hazards and conditions that they detect while driving. These products rely heavily on the accuracy of incoming data, particularly the driving speed limits within their embedded maps. Read More
Solution	Coyote used Dataiku Data Science Studio to develop an algorithm that leverages vast amounts of IoT-derived data. The algorithm segments roads into sections and analyzes patterns in each section, enabling Coyote to build a predictive model that estimates the speed limit of the road section. This machine learning process facilitated the detection of speed limit anomalies and enabled Coyote to estimate the global quality and reliability of the displayed speed limit. The collaborative functionalities of Dataiku DSS were crucial in this process, enabling employees with differing skill-sets to work together and promoting a widespread understanding of data mining, visualization, and smart data issues within the company. Read More Log in to view content
Contents

Technology Category

Analytics & Modeling - Predictive Analytics

Analytics & Modeling - Real Time Analytics

Applicable Industries

Transportation

Applicable Functions

Logistics & Transportation

Use Cases

Real-Time Location System (RTLS)

Predictive Maintenance

Services

Data Science Services

Challenge

Coyote, the European leader of real-time road information, faced a significant challenge in maintaining the accuracy of speed limit data within their embedded maps. This data is crucial for the functioning of their IoT devices and mobile applications, which warn drivers of traffic hazards and conditions. The company needed an automated, algorithm-based solution that could correct speed limit data and leverage the high volume of incoming data from their IoT devices to generate actionable insights and predictions. This also required instilling a data-driven approach within the company, where decisions are based on real-world data rather than standard analytics reports.

About Customer

Coyote is the European leader in real-time road information. The company was founded in 2005 and has 250 employees. It boasts 4.8 million users across Europe and generated a turnover of over €100 million in 2014. Coyote's products include IoT-based devices and mobile applications that enable users to warn other drivers of traffic hazards and conditions that they detect while driving. These products rely heavily on the accuracy of incoming data, particularly the driving speed limits within their embedded maps.

Solution

Coyote used Dataiku Data Science Studio to develop an algorithm that leverages vast amounts of IoT-derived data. The algorithm segments roads into sections and analyzes patterns in each section, enabling Coyote to build a predictive model that estimates the speed limit of the road section. This machine learning process facilitated the detection of speed limit anomalies and enabled Coyote to estimate the global quality and reliability of the displayed speed limit. The collaborative functionalities of Dataiku DSS were crucial in this process, enabling employees with differing skill-sets to work together and promoting a widespread understanding of data mining, visualization, and smart data issues within the company.

Impact #1	Speed limit reliability increased by 9% on analyzed datasets.
Impact #2	Automation of the speed limit correction process was achieved.
Impact #3	A global data-driven spirit was instilled within the company.

Impact #1

Speed limit reliability increased by 9% on analyzed datasets.

Impact #2

Automation of the speed limit correction process was achieved.

Impact #3

A global data-driven spirit was instilled within the company.

Benefit #1	Speed limit reliability increased by 9% on analyzed datasets.
Benefit #2	Automation of the speed limit correction process was achieved.
Benefit #3	Customer loyalty was increased.

Benefit #1

Speed limit reliability increased by 9% on analyzed datasets.

Benefit #2

Automation of the speed limit correction process was achieved.

Benefit #3

Customer loyalty was increased.

Download PDF Version

Overview

Anomaly Detection: How to Improve Core Product Accuracy and Efficiency with IoT

Operational Impact

Quantitative Benefit