Spatially Aware Model for Optimal Site Selection: Method and Application in a Greek Mobility Center

People’s perceptions about the appropriate location of mobility centers are modeled. A case study for the Municipality of Kalamaria in Thessaloniki, Greece, is presented. The results of a stated preference survey, which revealed residents’ perceptions about the optimal locations for these centers, as well as the spatial and other characteristics of the region, were used as inputs. Models were then developed and estimated: a linear regression model, a spatial simultaneous autoregressive (SAR) model, and a geographically weighted regression (GWR) model. The exploratory analysis used the respondents’ perception of optimal location as the response variable, and the explanatory variables are information about the population; points of interest; bus routes; average income; road network (i.e., length, number of lanes, and average speeds); and distances from stadium, pedestrian shopping zone, and paid parking facility. Results indicated that the selection of the appropriate location for a mobility center was influenced primarily by number of residents, number of points of interest, number of bus routes, average vehicle speed, and length of the road network. The GWR model seemed to fit the data best, whereas Moran’s I-test showed that the model solved the autocorrelation of the residuals, something that the other models [ordinary least squares (OLS) and SARmix] fail to accomplish. The model’s predicted values were localized (unlike OLS) and made the decision easier. The proposed method is generic (not system dependent) and can be applied to different optimal location selection schemes, especially those with similar demand characteristics (e.g., for automobile- or bike-sharing stations).

Language

  • English

Media Info

Subject/Index Terms

Filing Info

  • Accession Number: 01373942
  • Record Type: Publication
  • ISBN: 9780309223379
  • Report/Paper Numbers: 12-0597
  • Files: TRIS, TRB, ATRI
  • Created Date: Feb 8 2012 4:55PM