Abstract: Vast knowledge of construction machinery is brought to bear in the design and construction of an artificial intelligence machine that accepts telemetry data of uncertain origin and, through a set of decisions executed within the machine, and produces candidate telemetry channel identification data at its output. Candidate telemetry channel identification data may be considered sufficiently accurate when a confidence measure generated by the artificial intelligence machine meets a confidence threshold condition. Otherwise, the candidate telemetry channel identification data is returned to the artificial intelligence machine in which hypotheses are generated as to the identities of the received telemetry data.

Abstract: In some aspects, the techniques described herein relate to a method for object weight estimation, including: training an object identification model; receiving one or more images from a camera positioned on a machine engaged in a lifting operation; feeding each of the one or more images into the object identification model; receiving, from the trained object identification model, a predicted object corresponding to the object to be lifted by the machine and a probability that the predicted object corresponds to the object being lifted by the machine; and estimating a weight of the object based on at least the predicted object and comparing the estimated weight to lifting parameters associated with the machine when the probability is greater than a selected confidence threshold.

Abstract: Systems and methods for suggesting drop-box locations include collecting data from one or more sensors or databases, determining equipment location trends and asset usage trends in a geographic area based on the data, determining a customer opportunity score for one or more pieces of equipment in a geographic area, and, using an algorithm with a plurality of factors, combining the equipment location trends and asset usage trends with the customer opportunity score to output a geographic location for a drop-box. In some embodiments, the plurality of factors can include at least one or more customer distances from the geographic location.

Abstract: Techniques are provided that may include receiving sensor data from one or more machine sensors, the sensor data indicating machine failure. Informational data about current repair technician backlog, shop availability backlog, and parts availability are retrieved and inputted to a machine learning model for computing an estimated time to when the failed machine will be repaired. Customer data and dealer site data are retrieved and an estimated financial impact from a potential loaner machine is computed. The various computed and retrieved data are inputted to a decision tree algorithm, which computes an optimal solution for maintaining customer productivity. The network-accessible dealer site is notified of the optimal solution, which is intended for the customer.

Abstract: The present technology is generally directed to systems and methods for determining a customer-specific price for a replacement component of a machine. The present technology can include collecting, for one or more customers, data associated with the replacement component. For each of the customer(s), the present technology can further include determining one or more customer-specific factors; using the customer-specific factors(s) to determine one or more optimization metric(s); applying an optimization engine to the optimization metric(s) to determine a customer-specific adjustment; and inputting the customer-specific adjustment into a pricing system to generate a future price for the replacement component.

Abstract: A method for improving machine performance based on machine application identification can include training an application identification model. The method can also include receiving one or more images from a camera positioned on a machine performing an application, and feeding each of the one or more images into the trained application identification model. The trained application identification model provides a predicted application corresponding to the application being performed by the machine. The model also provides a probability that the predicted application corresponds to the application being performed by the machine. Application optimization parameters, based on the predicted application, are retrieved and distributed to the machine when the probability is greater than a selected confidence threshold.