Systems of methods for managing operations of a classification yard

Abstract

Methods and systems for managing operations of a classification yard. In embodiments a release speed, coupling speed, and/or a predicted speed at one or more points of a route along which a cut is being routed is determined. A set of event messages of events that occurred during the traveling of the cut is generated. Real-world measurements associated with an actual speed of the cut at the one or more points of the route are obtained. Coefficients associated with the predicted speed of the cut at the one or more points are autotuned based on the real-world measurements, a status of one or more devices used to route the cut is determined based, at least in part, on thresholding analysis applied to the real-world measurements, and the set of event messages is stored in an event log for subsequent replaying in a graphical user interface (GUI).

What is claimed is: 
     
         1 . A method of managing operations in a classification yard, comprising:
 determining, for a cut being routed to a destination train, one or more of a release speed, coupling speed, and a predicted speed of the cut at one or more points of a route along which the cut is traveling to reach the destination train, wherein each in a set of event messages represents an event that occurred during the traveling of the cut along the route to reach the destination train;   obtaining real-world measurements associated with an actual speed of the cut at the one or more points of the route;   autotuning one or more coefficients associated with the predicted speed of the cut at the one or more points of the route based on the real-world measurements associated with the actual speed of the cut at the one or more points of the route;   determining a status of one or more devices used to route the cut to the destination train based, at least in part, on thresholding analysis applied to the real-world measurements associated with the actual speed of the cut at the one or more points of the route; and   storing the set of event messages in an event log for subsequent replaying in a graphical user interface (GUI).   
     
     
         2 . The method of  claim 1 , wherein determining the one or more of a release speed, coupling speed, and a predicted speed of the cut at the one or more points of the route includes:
 generating the predicted speed of the cut at the one or more points using a production set of tuning parameters associated with the one or more points.   
     
     
         3 . The method of  claim 2 , wherein autotuning the one or more coefficients associated with the predicted speed of the cut at the one or more points of the route based on the real-world measurements associated with the actual speed of the cut at the one or more points of the route includes:
 estimating a candidate set of tuning parameters associated with the one or more points based on the real-world measurements associated with the actual speed of the cut at the one or more points of the route;   generating a set of backoffice predictions of the speed of the cut at the one or more points using the candidate set of tuning parameters;   comparing the predicted speed of the cut at the one or more points and set of backoffice predictions to determine which of the production set of tuning parameters or the candidate set of tuning parameters for the one or more points yields more accurate speed predictions; and   determining to replace the production set of tuning parameters for one or more points with the candidate set of tuning parameters in response to a determination that the candidate set of tuning parameters yields more accurate speed predictions.   
     
     
         4 . The method of  claim 3 , wherein estimating the candidate set of tuning parameters associated with the one or more points includes applying a regression algorithm to the real-world measurements associated with the actual speed of the cut at the one or more points of the route to obtain the candidate set of tuning parameters associated with the one or more points. 
     
     
         5 . The method of  claim 3 , wherein comparing the predicted speed of the cut at the one or more points and the set of backoffice predictions includes:
 calculating a production absolute value average difference between the predicted speed of the cut at the one or more points and the actual speed of the cut at the one or more points;   calculating a backoffice absolute value average difference between the set of backoffice predictions and the actual speed of the cut at the one or more points;   comparing the production absolute value average difference and the backoffice absolute value average difference to determine which one of the production absolute value average difference and the backoffice absolute value average difference is smaller;   determining that the production set of control parameters yields more accurate speed predictions for the one or more points than the candidate set of control parameters in response to a determination that the production absolute value average difference is smaller than the backoffice absolute value average difference; and   determining that the candidate set of control parameters yields more accurate speed predictions for the one or more points than the production set of control parameters in response to a determination that the production absolute value average difference is not smaller than the backoffice absolute value average difference.   
     
     
         6 . The method of  claim 1 , wherein the one or more points includes a hardware device, and wherein determining the status of the one or more devices based, at least in part, on the thresholding analysis applied to the real-world measurements associated with the actual speed of the cut at the one or more points of the route includes:
 generating a set of deviation metrics between a set of predicted measurements at the hardware device and a set of actual measurements at the hardware device; and   applying the thresholding analysis to the set of deviation metrics to determine a status of the hardware device.   
     
     
         7 . The method of  claim 6 , wherein applying thresholding analysis to the set of deviation metrics applying one or more of a set of differential rules to the set of deviation metrics, wherein the set of differential rules includes one or more of:
 a first differential rule specifying that the status of the hardware device is based on whether a threshold percentage of the set of deviation metrics are outside of a range defined by plus or minus a measurement threshold;   a second differential rule specifying that the status of the hardware device is based on whether a median or average of the set of deviation metrics is within a range defined by plus or minus an average threshold;   a third differential rule specifying that the status of the hardware device is based on whether a spread range of deviation metrics values within a middle percentage of the set of deviation metrics is less than a spread threshold, wherein the middle percentage of the set of deviation metrics is defined by a range of deviation metrics values including a top percentile threshold of the deviation metrics values in the set of deviation metrics and a bottom percentile threshold of the deviation metrics values in the set of deviation metrics; and   a combination differential rule that includes a weighted combination of the results of one or more of the first differential rule, the second differential rule, and the third differential rule.   
     
     
         8 . The method of  claim 1 , further comprising:
 obtaining the set of event messages from the event log;   generating, for each event message of the set of event messages, a visual representation of the event that occurred during the traveling of the cut along the route represented by each respective event message of the set of event messages;   generating a graphical schematic diagram of the classification yard, wherein the graphical schematic diagram includes a graphical representation of one or more components of the classification yard;   determining, for each event message of the set of event messages, a component of the one or more components of the classification yard with which each respective event message is associated; and   replaying the set of event messages in the GUI, wherein replaying the set of event messages includes overlaying each visual representation of an event represented by a respective event message onto the graphical representation of a component determined to be associated with the respective event message.   
     
     
         9 . The method of  claim 8 , wherein the graphical representation of the one or more components of the classification yard includes positioning each component of the one or more components on a location within the graphical schematic diagram of the classification yard corresponding to a generally relative physical location of each component of the one or more components on the physical layout of the classification yard. 
     
     
         10 . The method of  claim 8 , wherein the visual representation of the event that occurred includes one or more of:
 a color-based indication of the respective event;   a numerical indication of the respective event; and   a size-based indication of the respective event.   
     
     
         11 . A system for managing operations in a classification yard, comprising:
 at least one processor; and   a memory operably coupled to the at least one processor and storing processor-readable code that, when executed by the at least one processor, is configured to perform operations including:
 determining, for a cut being routed to a destination train, one or more of a release speed, coupling speed, and a predicted speed of the cut at one or more points of a route along which the cut is traveling to reach the destination train, wherein each in a set of event messages represents an event that occurred during the traveling of the cut along the route to reach the destination train; 
 obtaining real-world measurements associated with an actual speed of the cut at the one or more points of the route; 
 autotuning one or more coefficients associated with the predicted speed of the cut at the one or more points of the route based on the real-world measurements associated with the actual speed of the cut at the one or more points of the route; 
 determining a status of one or more devices used to route the cut to the destination train based, at least in part, on thresholding analysis applied to the real-world measurements associated with the actual speed of the cut at the one or more points of the route; and 
 storing the set of event messages in an event log for subsequent replaying in a graphical user interface (GUI). 
   
     
     
         12 . The system of  claim 11 , wherein determining the one or more of a release speed, coupling speed, and a predicted speed of the cut at the one or more points of the route includes:
 generating the predicted speed of the cut at the one or more points using a production set of tuning parameters associated with the one or more points.   
     
     
         13 . The system of  claim 12 , wherein autotuning the one or more coefficients associated with the predicted speed of the cut at the one or more points of the route based on the real-world measurements associated with the actual speed of the cut at the one or more points of the route includes:
 estimating a candidate set of tuning parameters associated with the one or more points based on the real-world measurements associated with the actual speed of the cut at the one or more points of the route;   generating a set of backoffice predictions of the speed of the cut at the one or more points using the candidate set of tuning parameters;   comparing the predicted speed of the cut at the one or more points and set of backoffice predictions to determine which of the production set of tuning parameters or the candidate set of tuning parameters for the one or more points yields more accurate speed predictions; and   determining to replace the production set of tuning parameters for one or more points with the candidate set of tuning parameters in response to a determination that the candidate set of tuning parameters yields more accurate speed predictions.   
     
     
         14 . The system of  claim 13 , wherein estimating the candidate set of tuning parameters associated with the one or more points includes applying a regression algorithm to the real-world measurements associated with the actual speed of the cut at the one or more points of the route to obtain the candidate set of tuning parameters associated with the one or more points. 
     
     
         15 . The system of  claim 13 , wherein comparing the predicted speed of the cut at the one or more points and the set of backoffice predictions includes:
 calculating a production absolute value average difference between the predicted speed of the cut at the one or more points and the actual speed of the cut at the one or more points;   calculating a backoffice absolute value average difference between the set of backoffice predictions and the actual speed of the cut at the one or more points;   comparing the production absolute value average difference and the backoffice absolute value average difference to determine which one of the production absolute value average difference and the backoffice absolute value average difference is smaller;   determining that the production set of control parameters yields more accurate speed predictions for the one or more points than the candidate set of control parameters in response to a determination that the production absolute value average difference is smaller than the backoffice absolute value average difference; and   determining that the candidate set of control parameters yields more accurate speed predictions for the one or more points than the production set of control parameters in response to a determination that the production absolute value average difference is not smaller than the backoffice absolute value average difference.   
     
     
         16 . The system of  claim 11 , wherein the one or more points includes a hardware device, and wherein determining the status of the one or more devices based, at least in part, on the thresholding analysis applied to the real-world measurements associated with the actual speed of the cut at the one or more points of the route includes:
 generating a set of deviation metrics between a set of predicted measurements at the hardware device and a set of actual measurements at the hardware device; and   applying the thresholding analysis to the set of deviation metrics to determine a status of the hardware device.   
     
     
         17 . The system of  claim 16 , wherein applying thresholding analysis to the set of deviation metrics applying one or more of a set of differential rules to the set of deviation metrics, wherein the set of differential rules includes one or more of:
 a first differential rule specifying that the status of the hardware device is based on whether a threshold percentage of the set of deviation metrics are outside of a range defined by plus or minus a measurement threshold;   a second differential rule specifying that the status of the hardware device is based on whether a median or average of the set of deviation metrics is within a range defined by plus or minus an average threshold;   a third differential rule specifying that the status of the hardware device is based on whether a spread range of deviation metrics values within a middle percentage of the set of deviation metrics is less than a spread threshold, wherein the middle percentage of the set of deviation metrics is defined by a range of deviation metrics values including a top percentile threshold of the deviation metrics values in the set of deviation metrics and a bottom percentile threshold of the deviation metrics values in the set of deviation metrics; and   a combination differential rule that includes a weighted combination of the results of one or more of the first differential rule, the second differential rule, and the third differential rule.   
     
     
         18 . The system of  claim 11 , further comprising:
 obtaining the set of event messages from the event log;   generating, for each event message of the set of event messages, a visual representation of the event that occurred during the traveling of the cut along the route represented by each respective event message of the set of event messages;   generating a graphical schematic diagram of the classification yard, wherein the graphical schematic diagram includes a graphical representation of one or more components of the classification yard;   determining, for each event message of the set of event messages, a component of the one or more components of the classification yard with which each respective event message is associated; and   replaying the set of event messages in the GUI, wherein replaying the set of event messages includes overlaying each visual representation of an event represented by a respective event message onto the graphical representation of a component determined to be associated with the respective event message.   
     
     
         19 . The system of  claim 18 , wherein the visual representation of the event that occurred includes one or more of:
 a color-based indication of the respective event;   a numerical indication of the respective event; and   a size-based indication of the respective event.   
     
     
         20 . A computer-based tool for of managing operations in a classification yard, the computer-based tool including non-transitory computer readable media having stored thereon computer code which, when executed by a processor, causes a computing device to perform operations comprising:
 determining, for a cut being routed to a destination train, one or more of a release speed, coupling speed, and a predicted speed of the cut at one or more points of a route along which the cut is traveling to reach the destination train, wherein each in a set of event messages represents an event that occurred during the traveling of the cut along the route to reach the destination train;   obtaining real-world measurements associated with an actual speed of the cut at the one or more points of the route;   autotuning one or more coefficients associated with the predicted speed of the cut at the one or more points of the route based on the real-world measurements associated with the actual speed of the cut at the one or more points of the route;   determining a status of one or more devices used to route the cut to the destination train based, at least in part, on thresholding analysis applied to the real-world measurements associated with the actual speed of the cut at the one or more points of the route; and   storing the set of event messages in an event log for subsequent replaying in a graphical user interface (GUI).