Heel-dominated treatment distribution among multiple perforation clusters is frequently observed in plug-and-perforate (plug-and-perf) stages, causing small propped surface areas, suboptimal production, and unexpected fracture hits.

A multifracture simulator with a novel wellbore-fluid and proppant-transport model is applied to quantify treatment distribution among multiple perforation clusters in a plug-and-perf operation. A simulation base case is set up on the basis of a field treatment design with four clusters. Simulation results show that the two toe-side clusters screened out early in the treatment and the two heelside clusters were dominant. The simulated proppant placement is consistent with distributed-acoustic-sensing observations.

The impact of different perforating strategies and pumping schedules on final treatment distribution is investigated. Two criteria are defined that quantify the proppant distribution and fracture area: the weighted average (WA) and standard deviation (SD) of the final fluid and proppant distribution, as well as the hydraulic surface area (HSA) and propped surface area (PSA) of the created fractures. An optimal plug-and-perf design is defined as one that minimizes the SD of the treatment distribution among perforation clusters, and maximizes the PSA.

Both perforating strategy and pumping schedule are found to affect the final treatment distribution significantly, and uniform treatment distribution is shown to create more PSA. Having fewer perforations per cluster was found to promote uniform fluid and proppant placement. Other helpful strategies include reducing the number of perforations near the heel and using small, lightweight proppant. The stress shadow effect is accounted for using the displacement discontinuity method (DDM) and was found to play a smaller role than perforation friction and proppant inertia in most cases.

An automated process is developed to optimize plug-and-perf completion design with multiple decision variables using a genetic algorithm (GA). Thirteen parameters are optimized simultaneously. The optimal design solution creates an almost even treatment distribution and more than doubles the PSA compared with the base case.

The multifracture model presented in this paper provides a way to quantify fluid and proppant distribution for any perforating strategy and pumping schedule, and provides more insight into the physics relevant to plug-and-perf treatment distribution. The perforation and pumping schedule recommendations presented in this paper provide directional guidance for the design of fracturing jobs with balanced treatment distribution and large PSA.