The Boeing Company

Liessman E Sturlaugson

James M Ethington

Eric Lee Nicks

Timothy Joseph Wilmering

Jason M Keller

Mark H Boyd

Nile Hanov

Seema Chopra

John M Boggio

James Schimert

Ameya Deepak Kamat

Daniel D Gilbertson

Ranjan Kumar Paul

Jeffrey Alan Schmitz

Harrison Kuo

Peter G Rhodes

Ali K Babool

Jan Irvahn

Ranjan K Paul

Christopher D Deits

Christopher David Deits

Rajkumar Srinivasan

An example method performed by a computing system for determining a maintenance interval for a subject aircraft configuration comprises obtaining sensor data reported by an electronic system of a population of the subject aircraft configuration. The method further comprises obtaining a failure mode definition that identifies a set of failure modes involving a component of the subject aircraft configuration. The method further comprises implementing a first predictive model to determine a first lifetime-probability distribution of a failure mode involving the component based on the sensor data. The method further comprises implementing a second predictive model that differs from the first predictive model to determine a second lifetime-probability distribution of a failure mode involving the component based on the sensor data. The method further comprises determining a maintenance interval for the component based on the first lifetime-probability distribution and the second lifetime-probability distribution.

Determining maintenance intervals using a combination of models

A method includes training a model to identify anomalous portions of a test component using training images and labels that indicate anomalous portions of training components within the training images. The method also includes compressing a source image of the test component to generate a first input image having a first resolution and making a first determination of whether the first input image indicates that the test component is anomalous. The method also includes making a second determination, for each section of a second input image, of whether the section indicates that the test component is anomalous. The second input image has a second resolution that is greater than the first resolution. The method also includes providing an indication of whether the first input image indicates that the test component is anomalous and providing an indication of whether the second input image indicates that the test component is anomalous.

Methods and systems for inspecting components

A method of categorizing natural language text is disclosed. The method includes selecting candidate keywords from a list of potential keywords based on the natural language text. The method also includes generating, using a classification machine learning model (MLM), a list of candidate categories based on the potential keywords. The method also includes generating, using a similarity comparison MLM, a similarity score based on the candidate categories and a set of pre-determined categories. The method also includes assigning a selected category based on the similarity score to the natural language text.

Machine learning for categorizing text

A method is provided that includes accessing a multivariate time series of flight data for an aircraft, and iteratively performing runs of genetic programming on groups of the sensors. A population of computer programs is randomly generated from a selected group of the plurality of sensors, and primitive functions selected from a library of primitive functions. The population is iteratively transformed into new generations of the population, and includes sub-rankings of the group of sensors based on a quantitative fitness determined according to selected fitness criterion. A ranking of the group of sensors from the sub-rankings of the group of sensors is produced. An aggregate ranking of the plurality of sensors is produced from the ranking of the group of sensors over a plurality of iterations. And the subset of sensors is selected from the aggregate ranking of the plurality of sensors, and according to selected optimization criterion.

Automated feature generation for sensor subset selection

A data processing system may include instructions stored in a memory and executed by a processor to categorize a plurality of systems into clusters using an unsupervised machine learning method to analyze repair data parameters of a historical dataset relating to the plurality of systems. The system may assign a repair forecast to each cluster, and may generate a system repair forecasting model using selected predictor variables, the historical data set, and the repair forecasts according to a supervised machine learning method. The selected predictor variables may correspond to a mathematical combination of operational data parameters in the historical dataset.

Machine learning based repair forecasting

A method includes receiving, by an equipment monitoring system (EMS), first training data and second training data from a plurality of sensors of an article over a first operating interval and over a second operating interval of the article, respectively, appending buffer data to the first training data, and the second training data to the buffer data to provide extended training data, and clustering the extended training data into a plurality of data clusters associated with operational states of the article. Subsequent to clustering, the EMS receives operational data associated with the plurality of sensors of the article over a third operating interval. The EMS determines a particular data cluster of the plurality of data clusters to which the operational data belongs, and communicates data indicative of an operational state of the article associated with the particular data cluster to a control system.

Method and system for reducing air-to-ground data traffic communicated from an aircraft

A vehicle maintenance scheduling and fault monitoring apparatus includes a vehicle system maintenance rules generation module and a vehicle system fault detection module. The rules generation module determines a correlation between pairs of precedent historical vehicle fault data, of historical time-stamped vehicle fault data, and a subsequent different historical vehicle fault data, and generates vehicle system maintenance rules based on the correlation determined for the pairs of the precedent historical vehicle fault data and the subsequent different historical vehicle fault data. The fault detection module monitors faults of the vehicle system, determines an imminent occurrence of a subsequent vehicle fault, based on application of the vehicle system maintenance rules to the plurality of time-stamped precedent vehicle fault data, and generates a maintenance report corresponding to the imminent occurrence of the subsequent vehicle fault so that proactive maintenance is performed on the vehicle system.

Apparatus and method for vehicle maintenance scheduling and fault monitoring

A maintenance interval adjuster and methods for improving accuracy of maintenance scheduling and changing a maintenance interval are presented. Scheduled maintenance data and unscheduled in-service maintenance data for a maintenance task are retrieved for a plurality of platforms. A distribution of lifetimes for the maintenance task in the scheduled maintenance data and unscheduled in-service maintenance data are analyzed for high variance or multiple modes. A number of conditions in sensor data of the plurality of platforms correlated to a length of the lifetimes for the maintenance task is identified, in response to identifying at least one of high variance or multiple modes in the distribution of lifetimes. The lifetimes are divided into a plurality of groups based on the number of conditions. A respective recommended maintenance interval is determined for each group of the plurality of groups based on respective lifetimes for the maintenance task of a respective group.

Adjusting maintenance intervals for individual platforms based on observable conditions

The disclosed examples relate to arranging graph data for display on a display device. One example provides, on a computing device, a method comprising determining one or more connected groups of nodes in directed acyclic graph data, for each connected group of nodes, determining a reachability from each node with no inputs to each of one or more nodes with no outputs to determine a plurality of initial node/terminal node pairs, and for each initial node/terminal node pair, determining a path from the initial node to the terminal node. The method further comprises initializing a grid based upon the determined paths, placing each node at a corresponding initial grid location, and placing edges between nodes based upon the determined paths to form an initial grid representation, modifying the initial grid representation via a cost function to form a modified grid representation, and outputting the modified grid representation for display.

Arranging directed acyclic graph data for display

Method and apparatus for detecting anomalous flights. Embodiments collect sensor data from a plurality of sensor devices onboard an aircraft during a flight. Feature definitions are determined, specifying a sensor device and an algorithm for deriving data values from sensor data collected from the device. Embodiments determine whether anomalous activity occurred during the flight using an anomaly detection model. An anomaly is detected including at least one of (i) a contextual anomaly where a data instance of a plurality of data instances is anomalous relative to a specific context, or (ii) a collective anomaly where two or more data instances are anomalous relative to a remainder of the plurality of data instances, even though each of the two or more data instances is not anomalous in and of itself. A report specifying a measure of the anomalous activity for the flight is generated.

On-board networked anomaly detection (ONAD) modules

A vehicle includes a first system, a first sensor, a second sensor, and a processor. The first sensor is configured to generate first sensor data indicative of an operational status of the first system. The second sensor is configured to generate second sensor data associated with an adaptive performance metric adjustment criterion. The processor is configured to, in response to determining that the second sensor data satisfies the adaptive performance metric adjustment criterion, determine, based on the second sensor data, an adaptive performance metric to be compared to the first sensor data to identify anomalous behavior of the first system. The processor is further configured to compare the first sensor data to the adaptive performance metric and to generate a maintenance alert in response to determining that the first sensor data fails to satisfy the adaptive performance metric.

Vehicle anomalous behavior detection

Methods include qualifying a qualification dataset of fatigue-related parameters calculated from data collected during a first group of flights of a first aircraft that experienced a replacement of a structural component during heavy maintenance. Methods include receiving a qualified selection of the fatigue-related parameters and verifying a verification dataset of the qualified selection of the fatigue-related parameters calculated from data collected during a second group of flights of a second aircraft that experienced heavy maintenance without replacement of the structural component. Methods include receiving a set of verified and qualified fatigue-related parameters and building a predictive model for structural repair during heavy maintenance with a training dataset of the verified and qualified fatigue-related parameters calculated from data collected during additional flights of the fleet. Hybrid feature selection systems also are disclosed.

Prediction methods and systems for structural repair during heavy maintenance of aircraft

Hybrid feature selection methods include methods of creating a predictive model for valve performance in a fleet of aircraft. Methods include qualifying a qualification dataset of valve-related parameters calculated from data collected during a first series of flights at least before and after a non-performance event of a valve. Methods include receiving a qualified selection of the valve-related parameters and verifying a verification dataset of the qualified selection of the valve-related parameters calculated from data collected during a second series of flights. Methods include receiving a set of verified and qualified valve-related parameters and building a predictive model for valve non-performance with a training dataset of the verified and qualified valve-related parameters calculated from data collected during additional flights of the fleet.

Hybrid feature selection for performance prediction of fluid control valves

Method and apparatus for detecting anomalous flights. Embodiments collect sensor data from a plurality of sensor devices onboard an aircraft during a flight. A plurality of feature definitions are determined, where a first one of the feature definitions specifies one or more of the plurality of sensor devices and an algorithm for deriving data values from sensor data collected from the one or more sensor devices. Embodiments determine whether anomalous activity occurred during the flight using an anomaly detection model, where the anomaly detection model describes a pattern of normal feature values for at least the feature definition, and comprising comparing feature values calculated from the collected sensor data with the pattern of normal feature values for the first feature definition. A report specifying a measure of the anomalous activity for the flight is generated.

Growing community of inventors

St. Louis, MO, United States of America

Liessman E Sturlaugson