1538 2.RESEARCH METHODOLOGY This research uses Adaptive Machine Learning Model to ensure practical usability and sustained performance in dynamic operational environments, that continuously updates its predictive capability based on recent telemetry data. The model is designed to support operational decision-making by providing realtime and near-real-time recommendations that reflect current road, vehicle, and operational conditions. In open-pit coal mining operations, haul truck productivity represents a primary performance indicator, as hauling typically accounts for the largest proportion of operational cost and time. Productivity is defined as the amount of material transported per unit time and is directly influenced by payload capacity, haul distance, road conditions, and operator driving behavior. The productivity of a haul truck can be mathematically expressed as: L= S P D where Ldenotes truck productivity (tons/hour), Pis the effective payload per cycle (tons), and SDis the total cycle time (hours). The illustration of truck’s cycle is described in below: Figue 1. Truck’s Hauling Cycle Truck’s hauling cycle starts from the arrival of truck in front loading area until travels to disposal area and return to front area. When there are any other haulers in front when a hauler arrive, it needs to queue until the precedent hauler has completed the loading process. The total cycle time is decomposed into several sequential operational phases: SD =SHC5, +SF5IH +S,I-& +S12>I1$ +S,2H5' where: SHC5, is the loading time at the shovel, SF5IH is the loaded travel time from the loading point to the dumping location, S,I-& is the dumping or unloading time, S12>I1$ is the empty return travel time, and S,2H5' includes operational delays such as queuing, speed fluctuations, and unnecessary braking. Among these components, SF5IH, S12>I1$, and S,2H5'are strongly influenced by driver behavior and therefore represent the primary optimization targets of the proposed Automatic Operator Guidance System (AOGS). Haul and return times are directly governed by the vehicle speed profile along the haul road. Let : represent the haul road length; the haul time can be expressed as: S F5IH =` ¡( 1 () N < 4( where ¡(()is the truck speed as a function of road position (. Variations in speed due to inconsistent driving behavior, excessive braking, or conservative speed selection significantly increase the integral value of SF5IH, thereby reducing overall productivity. By reducing speed variance and maintaining operation near the optimal speed envelope, the guidance system minimizes travel time without compromising safety constraints. 2.1 Target Variables The objective of the Automatic Operator Guidance System is to determine an optimal speed profile that minimizes haulage time while satisfying safety, mechanical, and environmental constraints. This problem is formulated as a multi-objective optimization:
RkJQdWJsaXNoZXIy MTM0Mzk2