An iron-scintillator sandwich calorimeter is being developed for inclusion as part of a second detector at the upcoming Electron-Ion Collider. The detector would provide excellent muon and neutral particle identification through the use of time of flight information collected by the calorimeter. The timing information would also enable energy reconstruction and longitudinal segmentation. The present study focuses on the optimization of these aspects of the detector. Detector simulations are accomplished with the DD4HEP framework, utilizing GEANT4. Generative AI is used to implement a fast timing parameterization to allow for simulating particle detection without the simulation of tens of thousands optical photons. Machine learning techniques are utilized to identify particles and predict energy. Multi-Objective Bayesian Optimization is performed to identify the detector parameters that maximize energy resolution and particle identification accuracy. Measuring the Thermal Sunyaev-Zel'dovich Effect in Galaxy Clusters with ACT

Abstract profile. Full document pending author claim.