Combining Stochastics and Analytics for a Fast Monte Carlo Decay Chain Generator

TL;DR

This paper introduces a hybrid analytic and stochastic algorithm for efficiently simulating radioactive decay chains, reducing computational costs while accurately modeling decay sequences and correlations without post-processing.

Contribution

It presents a novel combined analytic/stochastic method for fast, accurate decay chain simulation starting from arbitrary source ages, eliminating the need for post-processing.

Findings

Significantly reduces simulation time for decay chains.

Maintains accurate time and position correlations.

Avoids extensive post-processing steps.

Abstract

Various Monte Carlo programs, developed either by small groups or widely available, have been used to calculate the effects of decays of radioactive chains, from the original parent nucleus to the final stable isotopes. These chains include uranium, thorium, radon, and others, and generally have long-lived parent nuclei. Generating decays within these chains requires a certain amount of computing overhead related to simulating unnecessary decays, time-ordering the final results in post-processing, or both. We present a combination analytic/stochastic algorithm for creating a time-ordered set of decays with position and time correlations, and starting with an arbitrary source age. Thus the simulation costs are greatly reduced, while at the same time avoiding chronological post-processing. We discuss optimization methods within the approach to minimize calculation time.