The first-order model is the most widely utilized model for chlorine decay due to its having only one parameter and an analytical solution. While variable reaction coefficient type models have higher accuracy and wider application range, their widespread uses are hindered by their complexity and the non-existence of an analytical solution. The objective of this study was to develop a variable parabolic reaction coefficient model to simulate and predict chlorine decay in bulk water. The decreasing reactivity of the reacting agents is included in a variable coefficient, which decreases with increasing consumption of the chlorine-reactive species concentration. The model includes minimal parameters that must be calculated, and an analytical solution was derived. Experimental data, including chlorination with different initial chlorine concentrations or temperatures, rechlorination, and water mixing, were utilized to evaluate the accuracy of the variable parabolic reaction coefficient model under different conditions. The relationship between the parameters and temperature was established utilizing the Arrhenius equation. On the basis of assumptions that the chlorine consumed by reactions with the pipe wall does not decrease the variable rate coefficient, this study subtracted the instantaneous aggregate chlorine wall consumption from the total chlorine consumption, so that the parameter values in the VPRC model (derived from laboratory decay tests) can be used in the system model. For output water at the entrance of the WDN with unknown initial chlorine concentration and unknown decay duration from disinfection to entry into the water distribution network, an effective method to determine the model parameter is proposed.