Electricity service providers (ESP) worldwide have increased their interest in the use of electrical distribution, transmission, generation, storage, and responsive load resources as integrated systems. Referred to commonly as "smart grid," their interest is driven by widespread goals to improve the operations, management and control of large-scale power systems. In this thesis I provide research into a novel agent-based simulation (ABS) approach for exploring smart grid system (SGS) dispatch, schedule forecasting and resource coordination. I model an electrical grid and its assets as an adaptive ABS, assigning an agent construct to every SGS resource including demand response, energy storage, and distributed generation assets. Importantly, real time is represented as an environment variable within the simulation, such that each resource is characterized temporally by multiple agents that reside in different times. The simulation contains at least as many agents per resource as there are time intervals being investigated. These agents may communicate with each other during the simulation, but only agents assigned to represent the same unique resource may exchange information between time periods. Thus, confined within each time interval, each resource agent may also interact with other resource agents. As with any agent-based model, the agents may also interact with the environment, in this case, containing forecasted environment, load and price information specific to each time interval. The resulting model is a time-independent global approach capable of: (1) capturing time-variant local grid conditions and distribution grid load balancing constraints; (2) capturing time-variant resource availability and price constraints, and finally, (3) simulating efficient unit-commitment real-time dispatches and schedule forecasts considering time-variant forecasted transactive market prices. This thesis details the need for such a system, discusses the form of the ABS, and analyzes the predictive behavior of the model through a critical lens by applying the resulting proof-of-concept simulation to a set of comprehensive validation scenarios. The resulting analysis demonstrates ABS as an effective tool for real-time dispatch and SGS schedule forecasting as applied to research, short-term economic operations planning and transactive systems alike. The model is shown to converge on economic opportunities regardless of the price or load-forecast shape and to correctly perform least-cost dispatch and schedule forecasting functionality.