CUBE LAND

CUBE Land is a powerful economic land-use forecasting software, which allows the user to model the interaction between real estate markets and transportation systems.

Image 1 - Model Structure

CUBE Land can be added to any existing CUBE application as a new program step in CUBE Base, which also provides the embedded ArcGIS interface for defining model inputs as well as mapping and reporting model outputs.

The key benefits of using CUBE Land as the tool for land-use modelling are:

• Possibility to efficiently integrate the land-use model with a CUBE transportation model
• High program flexibility for treating transport’s effects on land-use
• High program flexibility in the definition of the drivers of households and firms location and in the segmentation of the agents and real estate market
• High program flexibility in the definition of the zoning system level of detail
• Tidy management of input/output data and clear visualization of the full modelling process
• Direct management of different scenarios within the CUBE Base logic

CUBE Land is based upon Dr. Francisco Martinez’s highly regarded and innovative MUSSA II model (Martinez, F. “Towards a Land-Use and Transport Interaction Framework” in Hensher and Button, Handbook of Transport Modelling, Elsevier).

CUBE Land’s  main features can be summarized in the following five aspects:

• CUBE Land uses basic economic principles of real estate market equilibrium
• CUBE Land allocates households and employment by type from control totals to sub-regional zones
• CUBE Land forecasts commercial and residential development by type and zone
• CUBE Land features a real estate supply model that can reflect fixed bounded, or constrained supply assumptions
• CUBE Land calculates real estate prices based upon the amount the highest bidder would be expected to pay in an auction (bid-rent theory)

Image 2 - LUTI representation

CUBE Land provides a streamlined interface for developing transport and land-use interaction models (LUTI).

CUBE LAND – methodology

CUBE Land is based on a “bid-choice” framework, combining the “bid-rent” theory with discrete choice models. CUBE Land consists of three subsequent and interconnected models:

• Demand model, representing the consumers behavior (household and firms)
• Supply model, representing the real estate developers behavior
• Equilibrium model (rent model), determining the rent values and verifying the equilibrium conditions

A real estate market equilibrium between suppliers and consumers is determined, subject to bounds, constraints, restrictions, and policy assumptions.

CUBE Land incorporates a clear behavioral logic to determine the equilibrium: real estate properties are occupied by the household/firm willing to pay the highest price, whilst developers maximize profits when deciding what type of buildings to provide.

Location externalities and economies of scale are incorporated via a feedback within the equilibrium loop, simulating the interaction between consumers and suppliers.

Image 3 - Relations between/within models and feedback loop

CUBE LAND – LUTI

CUBE Land can be configured to implement any of a number of widely-known approaches for integrating travel demand and land-use model components. You can produce an equilibrium forecast based on automatic feedback between land-use and transportation models, introduce time lags between sub-models, or manually control the interaction between model components.

Household and firm bid/rent functions consider transportation service quality:

• Households consider “accessibility” as the possibility to access desirable destinations in the area (active); and,
• Firms consider “attractiveness”, that is how easy it is for customers to access the place of business (passive).

 Both of these aspects are marked-segment specific, therefore depending on the household type (income and size) and on the firm industry category.

Accessibility and attractiveness can be measured with any user specified formulation from the transport model skim matrices and other data.

CUBE LAND – Input Data

The input data structure and formulations are flexible in CUBE Land, depending on the specific model characteristics and data availability. With input and configuration files, you specify the required information and the elements that define the city and market that CUBE Land simulates. Specifically, you define and categorize the zones, the agents and the properties:

• Zonal characteristics:
  • Zonal accessibility and attractiveness
  • Total area by land use category
  • Average household income
  • Zoning policies and other restrictions
• Market-segmented:
  • Residential (Households) stratified by size, income, etc.
  • Non-Residential (Firms) stratified by industry, size, etc.
• Real estate unit characteristics:
  • Average lot size
  • Average floor-space
  • FAR (Floor Area Ratio)
  • Average monthly rent, etc.

Zonal input data can be stored in attribute tables of polygon GIS layers. The level of resolution depends on the zone’s arbitrary size (parcels, blocks/block groups, Traffic Analysis Zones – TAZ, census tracts, zip codes, etc.). Other than geodatabases, input formats are TEXT Files and DBF tables.

CUBE LAND – Outputs

The outputs from CUBE Land are DBF files that can be easily post-processed within the CUBE environment to calculate additional indicators and display them to communicate more effectively by visualizing outputs in CUBE GIS or ArcGIS directly.

Typical outputs at the equilibrium condition are:

• Number of occupied real estate properties (e.g. housing units) by type in each zone
• Households and firms by segment, zonal location and real estate property type
• Zonal endogenous attributes at equilibrium (location externalities): e.g. average household income by zone, built area by land-use type, etc.
• Variations in number of households/population and jobs for each zone due to changes in the transport network or testing policies
• Rents by real estate type and zone (simulated relative property values)
• Bids of each consumer type (households and firms) for each type of real estate and zone

CUBE LAND – Applications

At the base of any transport model, there is the underline concept that transport needs are induced by the need to perform activities. Transport demand and supply depend on the location of activities and land-use within the study area. Vice-versa, the transportation system does influence the activity system and the land-use. This creates a cycle loop between land-use and transport costs.

CUBE Land, integrated in the CUBE environment, is the tool developed to analyze these interactions, with the following three main objectives:

• To improve transport modelling, obtaining better forecasts in terms of mobility, incorporating land-use changes in transport demand modelling
• To simulate the interactive loop between the transport system and the activity system, incorporating the concept of accessibility
• To analyze sociological, environmental and economic impacts of land-use policy measures and transport planning

With CUBE Land you can:

• Move from a concept of “mobility planning” to a concept of “accessibility planning” at urban and regional level
• Evaluate to what extent transport projects induce urban development and measure to what extent household and job activities location choices depend on the accessibility provided by the transportation system:
  • Test alternative networks to see effects of transportation projects on land use pattern
  • Analyze the variation in household and job activities distribution due to major reconstruction/upgrade of the existing road network
  • Evaluate the impact of major public transport infrastructure projects, such as a new underground railway, in terms of modal split from private to public transport, incorporating the changes in land-use
• Test alternative land use policies to see effects on transportation performance
• Include the impact of regulatory policies and monetary incentives (subsidies and taxes) on agents location choices (e.g. encouraging low-income housing)
• Forecast increased land values due to infrastructure projects (value capture calculations)

 

 

Image 4 - Household and Firms variations due to the opening of a new underground line (Milan - Italy)

 

CUBE Land has been successfully applied and tested around the world to analyze several different policies and projects. Here are some examples:

• Chile: evaluation of three mega-projects in Santiago (Transantiago BRT, Cerillos Airport and Central Ring) with the objective of obtaining the optimal allocation of subsidies (minimum transport cost for schools, emissions & CO2 permit trading)
• Thailand: activity based model developed for the Regional City of Phitsanulok to develop a road improvement plan for short and mid term (2015-2020)
• USA: development of a fully functioning integrated land-use and transport trial model for the Montgomery Area in collaboration with Auburn University
• USA: land-use and transport integrated model based on an extensive dataset to draft 2040 LRTP forecasts for the Metropolitan Council of the Twin Cities (MN)
• USA: large-scale regional model with 531 land-use zones (16.5 million people) for the Los Angeles Region, continuing uses for strategic planning and housing policy analysis purposes
• South Korea: large-scale model for entire Seoul metropolis using CUBE Land auto-calibration tools
• USA: CUBE Land model developed for the Louisville metropolitan region to study land-use effects of the proposed Ohio River Bridge projects
• Italy: land-use trial model for the Milan metropolitan area to study the impact of major public transport projects and land use policies, in collaboration with Milan University