Neutron-induced gamma-ray spectroscopy measurements originate from inelastic neutron scattering and thermal neutron capture of chemical elements excited by neutrons; they are widely used to quantify in situ elemental and mineral compositions of rocks in boreholes. Traditional methods of interpreting neutron-induced gamma-ray spectroscopy measurements neglect layer-boundary and layer-thickness effects in the estimation of elemental and mineral compositions. Such effects can cause significant averaging of true layer properties in thinly bedded formations or in formations penetrated by high-angle/horizontal (HA/HZ) wells. Reliable measurement interpretation must therefore begin with the development of a fast and accurate forward-simulation model that explicitly incorporates measurement physics as well as borehole, tool, and formation geometry. Numerical Monte Carlo methods accurately reproduce averaging effects but are extremely time consuming and impractical for use in routine spectroscopy interpretation. We have developed a rapid and accurate numerical method that uses spatial coupled neutron-gamma-ray sensitivity functions to account for environmental and 3D effects of formation porosity, fluids, dipping beds, thin beds, and arbitrary well trajectories in the simulation of elemental and mineral compositions. Simulations are performed assuming a logging-while-drilling spectroscopy tool with a 14-MeV pulsed-neutron source. We benchmark results obtained with the rapid simulation method against rigorous Monte Carlo spectroscopy calculations for synthetic conventional and unconventional thinly bedded reservoirs penetrated by vertical and HA/HZ wells. Fast simulations are obtained in approximately 1e-6 the time required for Monte Carlo simulations, with an average difference of less than 5% between them. Similar to Monte Carlo simulated logs, fast simulated logs accurately reproduce geometric effects on measurements affected by shoulder-bed averaging of layer elemental and mineral compositions. The fast-forward-simulation method facilitates the development of efficient inversion-based spectroscopy interpretations, which could mitigate geometric effects that mask true layer elemental and mineral compositions.