Machine learning resources

 

Intelligent Pattern Recognition for Operations

Falkonry accelerates the continuous improvement of production operations through intelligent pattern recognition in time-series data, enabling improvements in uptime, quality, yield, and efficiency. Falkonry uses the knowledge of resident Subject Matter Experts (SMEs) to provide context to its data-driven analysis, bringing insight and continuous improvement to industrial processes.

Read More

Batch Manufacturing with Falkonry

Specialty chemical manufacturing is a multi-step, complex process, with demanding yield requirements. The quality of each individually produced batch is dependent on the characteristic of the starting raw materials and process parameters at each production step. Long production cycles of high batch volumes mean that unacceptable variations in quality outcomes can rapidly accumulate lost time, material waste, scrap cost, and lost revenue. Raw material variation combined with process metrics in multi-step chemical production processes can predict the final quality of production batches. The challenge is to predict finished batch quality at various points of the production process.

Read More

Process Manufacturing Solution with Falkonry

Solar cell manufacturing is a high speed, multi-step, complex process, with demanding yield requirements. The quality of each individually produced solar cell is dependent on the raw material characteristic of the wafer and the process parameters of each production step. High production volumes mean that unacceptable variations in quality outcomes can rapidly accumulate lost time, material waste, scrap cost, and lost revenue. Finished solar cells are classified by quality grade. Raw material variation at the beginning of the multi-step solar cell production process can predict the final quality of individual solar cells. The challenge is to predict finished cell quality before production begins.

Read More

IT Performance Monitoring with Falkonry

Content distribution networks rely on accurate configuration files to direct HTTP requests to servers containing with the content being requested. But content is changing all the time: new content is being added, old is being updated, and expired is being removed. So configuration files are constantly changing. When errors are introduced in the configuration file itself, or exist because content is not where it is supposed to be, losses occur: customers do not receive the con- tent they seek causing sales and service to suffer. In today’s fast-paced world, poor service often translates to lost revenue and shifting of customers to competitors.

Read More

Converting Operations Solution with Falkonry

Converting operations depend on consistent and reliable machine operation to meet the demanding quality and productivity requirements of a profitable business. Employing advanced technology to extract more business value from manufacturing operations is becoming critical as manufacturing expertise ages out, plants depend on automation, and gains in production quality and operational effectiveness become market differentiators.

Read More

Falkonry Technical Overview

This paper describes a general-purpose system and an associated approach for condition recognition from time series data, and it discusses how this system and approach can be applied to a wide array of operations management solution needs. There are many operational situations where sensor data is available and condition understanding is desired, but where constraints limit the application of machine learning or other techniques – constraints could include financial and time constraints, lack of behavioral models, lack of data science resources, or lack of labeled training sets. The system and approach described in this paper support pragmatic condition recognition in the presence of these constraints. The described system includes implementations of feature learning (with a focus on temporal pattern recognition), clustering, and classification techniques integrated to support unsupervised and semi-supervised learning for diverse sets of multivariate signal data. A continuous learning approach is employed where, initially, only unnamed recurring patterns might be recognized through unsupervised learning.

Read More

Human Activity Tutorial

AI for Live Operations

Simple Batched Window Tutorial

Wheel Health Tutorial

Webinar: Using Patterns in Real-Time Data for Operational Excellence

May 17, 2017 11:00am PT | Dr. Nikunj Mehta, Falkonry

Webinar: Adding Falkonry to PI: A Lab in 50 Minutes

April 25, 2017 7:30am-8:30am PST | Dr. Greg Olsen, Falkonry

Lab: Adding Pattern Recognition Capabilities to Your PI System with Falkonry

Mar 20, 2017, 9am-12noon | Dr. Greg Olsen, Falkonry

How to Extend Splunk with an AI Assistant for Pattern Recognition

Sep 28, 2016, 3:30-4:15 pm | Dr. Greg OIsen, Falkonry