550 / 2019-02-28 21:43:59

Evaluation of sustainability for regeneration of old housing complex with household based urban micro-simulation model

micro-simulation, household location choice

全体主题 > Simulation

In 1960’s and 1970’s, housing demand in Japan was rapidly increased in accordance with formation of new families of baby boomers. Huge housing complex called new towns have been developed in everywhere. These new towns have changed into “old towns” after 40 to 50 years since developed. Several problems, deterioration of houses, decreasing population, withdrawal of commercial shores and public services and decline in the Quality of Life (QoL) are arising in the old towns. Regeneration plans are under planning to overcome the problems in many old towns. For making these residential places sustainable, it is essential to induce new residents, especially young generation who will have more children, to choose there as living places and live well in the long time as well as improving QoL for existing residents. Therefore it is necessary to consider how many and what types of households to move in and the change in family structure of households who live in the place now. The objective of this study is to forecast and evaluate sustainability of the residential place with household micro-simulation model named “HUMS (Household based Urban Micro-Simulation model)” 1) which we have developed. In this study, Hishino housing complex is selected as a target area for evaluation with this model. Hishino housing complex is located in Seto city of Japan. The place was developed in late of 1960’s and the majority of residents at that time are young generations. After 50 years since developed, the residents are getting older and most of their children went out their home town. So, the rate of elderly people is high (over 65 years: 36.9%). The total population is 13,113 as of 2015 which is decreasing after 1984 when the population was at the peak, 21,746.
Our micro-simulation model is divided into three parts, (1)Database, (2)Life events generation model, (3)Housing types and location choice model. Database is divided into personal and spatial database. The personal database includes all micro-data relating to the attributes of individuals including ID number of individual, households and residential zone, age, sex, marital status, number of people in the household, number of children, job status, type of house and household income. In the existing study1), the initial micro-data of residents for the base year is estimated from probabilities obtained from a sample micro-data set obtained from questionnaire survey. In this study, initial micro-data was estimated from data of national census which is one of representative open data. Probabilities of generation life events including aging, death, marriage, birth and relocation were estimated with several types of open data. Location choice model is divided into two level, upper and lower model. The former is modeled for estimating household location choice in 18 residential zones of Seto city. The latter is modeled for estimating household location choice in 10 zones within Hishino housing complex. Multinominal logit model is used to estimate both models. Parameters of these models can be estimated with high accuracy. Housing type choice model is also estimated for residents in Hishino housing complex.
Change of household structure and population distribution is simulated with estimated micro-simulation model for next 10 to 20 years for the case of BAU and with regeneration project such as improving accessibility to public transport or commercial facilities. At the conclusion, the results of evaluation of sustainability of the old housing complex estimated with our micro-simulation model and usability of the model.