TRB 2016 Blue Ribbon Committee
16th National Transportation Planning Applications Conference

Optimal Land Use Allocation for Sustainable Mixed Use Developments

Corresponding Author: Chang-Jen Lan, Town of Jupiter

Presented By: Chang-Jen Lan, Town of Jupiter


This study aims to seek the optimal land use allocation by maximizing the total profit from the commercial rentals and real estate sales at the mixed use development, while subjecting the overall external traffic impact to maintain adequate capacity and level of service on the adjacent roadway network for sustainability. Trip internalization is one of the key factors in determining optimal allocation. The degree of patron interaction among different uses, including office, retail, restaurant, residential, hotel and theatre, will depend on the size of leasable floor area, accessibility, synergy among uses, external attraction, etc. However, the existing state-of-the-art methods, including ITE and NCHRP 684, often yield inaccurate estimates due to lack of explanatory variables characterizing the site-dependent internalization behavior. In particular, the NCHRP 684 consistently overestimates the internal capture rates based on a questionable behavior assumption, which will impose financial burden on government agencies to rectify deficient traffic mitigations or the public will suffer from more congestion otherwise.

This study develops a systematic methodology to more accurately estimate and predict trip internalization at mixed-use developments. Coupled with such accessibility measures as use-specific production/attraction trips and synergies in the gravity model for quantifying the internal trip distributions among uses, the Dirichlet-multinomial distributional model is embedded in the generalized linear model framework for estimation. In addition, the weighted entropy and external attraction variables are used to enhance the explanatory power for the site-dependent variation in trip internalization. The estimation results indicate that, except the distances among uses, all the variables are statistically significant. The validation against cordon counts shows the overall internalization rates predicted from ten study sites were substantially improved over the existing methodology.

Lastly, since taking up significant footprint, the parking space needs to be specifically factored into the allocation plan. The space requirements for different uses are calculated based on the time-of-day utilization factors in a shared parking estimation routine. These procedures are embedded in the constrained optimization program to generate the optimal land use plan that can be provided to traffic engineers and land developers without the need for tedious trials and errors. To be useful to practitioners, a user friendly spreadsheet program, called TRIPAL (TRip Internalization, Parking and Allocation of Land), has been developed by the author for distribution as a sketch planning tool.


Discuss This Abstract