If adding 5% DCO to the Coker feed reduces the fouling rate, then the problem is likely an incompatible resid blend. DCO is very aromatic and a good solvent for asphaltenes, so blending DCO into the Coker feed will help keep the asphaltenes in solution and limit the coke formation in the heater.
When I have seen resid incompatibility the heater fouling occurs in the upper radiant section and the shock tubes of the convection section. Normal Coker heater fouling is typically in the lower radiant section of the heater. Where in the heater are you seeing this rapid fouling?
Asphaltene precipitation occurs when the solvent power of the resid is insufficient to keep all the asphaltenes in solution. Blending aromatic, high asphaltene resids with paraffinic, low asphaltene resids can result in significant asphaltene precipitation, depending on the proportions and the blending order. If resid blending is done by blending crudes together, then the order does not matter. The overall solvent power is the key to preventing the asphaltenes from coming out of solution early.
Just because the crudes are compatible at desalter temperatures does not mean that the resids will be compatible at the elevated temperatures in the Coker heater. Managing the ratios of aromatic to paraffinic resids is probably the most practical method of managing this fouling. I am not aware of a definitive test that predicts this fouling.