Biasing an environmental gamma-ray scattering simulation

I have a small detector under a very large pipe carrying various gamma-emitting gasses. I am trying to simulate the detector’s response to gamma-emitters passing through that pipe, not just the direct gamma-ray component but scattering off of the various support structures and equipment in the area.

I have the geometry worked out as well as a generator which randomly pics a point in the pipe and direction. Since I want to capture the detector behavior for gammas scattering off the roof, the floor, the pipe, the pipe support structure, etc before entering the detector, it seems to me that I would need to simulate many many events to achieve a reasonable response that captured the scattering term as well as the “mostly direct” term.

I write this hoping for some guidance in choice of biasing for this system. Currently the only biasing method I have read about in the guide for application developers that seems relevant is the Reverse/Adjoint Monte Carlo since the pipe is large and the detector is small. I am sure I am missing possibilities that might be simpler or less error prone and would love any suggestions and hints.