PARKING DEMAND MODEL FOR TERTIARY EDUCATION CENTER
KUNJAN H.M
Research Assistant, Faculty Of Civil Engineering,
Universiti Teknologi MARA, 40450 Shah Alam, Selangor, Malaysia.
TEY
L.S
KUAN W.K
Lecturer, Faculty Of Civil Engineering,
Universiti Teknologi MARA,
40450 Shah Alam, Selangor, Malaysia.
email :
teylisian@yahoo.com
ABDUL RAHMAN M.Y.3
Associate Professor, Faculty Of Civil Engineering,
Universiti Teknologi MARA,
40450 Shah Alam, Selangor, Malaysia.
INTRODUCTION
In the major cities of Malaysia, transportation system is always acting as a link to deliver goods from place to place and allows individual traveling from a place to his destination. It also forms as a medium for the daily social activities. The transportation system can be in the mode of air, water and land. In land transportation system, vehicles that moving on the roads need to stop regardless of the trip purposes. Therefore, parking facility is essential for temporarily keeping the vehicles so that the drivers are able to carry out their trip purposes. From the records of Road Transport Department (Jabatan Pengangkutan Jalan, 2006) there was an increment of new registered vehicles from 519,830 in year 1990 to 932,363 in year 2002 which gave 80% of increment. Therefore, improper planning for the allocation of parking spaces in future will worsen the existing traffic congestion problems which has been a ‘tumour’ to the local council in urban area. The role of local council as a planner become crucial since over supply of parking space will incur great expenses on construction and land and yet it will encourage vehicle ownership and usage. But an acute shortage of parking space would create traffic congestion as motorists searched or waited for parking especially in the urban area. The objective of this study is to develop a mathematical model for estimating vehicle parking demand for the category of land use of tertiary education center.
RESEARCH METHODOLOGY
The parking phenomenon is based on the law of supply and demand. Supply is defined as the total number of spaces available within a selected study area whereby demand refers to the total parking accumulation and illegal parking at a given time duration. In this study, Shah Alam campus of Universiti Teknologi MARA has been selected as the study areas and only academic departments being included. The areas solely for administration are excluded. The five selected study areas include:
1. Faculty of Architecture, Planning and Surveying (FSPU).
2. Faculty of Information technology and Quantitative Science (FTMSK).
3. Faculty of Civil Engineering, Faculty of Mechanical Engineering, Faculty of Electrical Engineering, Faculty of Chemical Engineering, Faculty of Medicine, Faculty of Pharmacy, two libraries – PTAR III and PTAR IV and Chancellery (S&T).
4. Faculty of Sport Science and Recreation, Faculty of Science Management and Policy Studies and Faculty of Communication and Media Studies (FSSR).
5. Faculty of Art and Design (FSS).
Cordon survey is adopted for collecting the vehicles parking demand data continuously from 7:30 am to 5:30 pm. In this study, vehicles refer as cars and vans only. Data collection for parking demand was conducted between December 2004 – September 2005 during semester II of session 04/05 and semester I of session 05/06. The selection of the time period is to reflect daily operating hours and there was no special event in the campuses during the sampling days. The information of floor areas in this study was collected from the Property Management Unit of the Development Division of Universiti Teknologi MARA (Unit Pengurusan Harta, 2006). The floor areas for each studied area only covered for administrative and academic purposes.
According to Maintenance Unit (Pejabat Penyelenggaraan), UiTM Shah Alam, the provision of number of parking spaces is based on the guideline published by the local council, Majlis Bandaraya Shah Alam, entitled “Panduan Permohonan Kelulusan Dan Piawaian Perancangan” (1995) which determined by three variables: Numbers of Staff, Students and Visitors for the tertiary education center. In the guideline, one parking space is provided for every two administrative or academic staff and in additional one parking space is for every ten students. Parking space for visitors is assumed as 10% of total parking spaces provided for staff and students. The guideline is published in 1995 and this may not be adequate for estimation now since the car ownership might be increased. A new guide is required especially for future development; for example a new faculty, in campus.
Regression analysis was used to develop a model that relates the parking demand with a specific independent variable. The values for parking demand were obtained through the daylong cordon survey from each study areas and the floor areas information was obtained from the Property Management Unit. In each site, only the highest accumulated parking demand within a day is used in the analysis. In this study, floor area is adopted as the independent variable instead of population of staff and students because the population varies throughout the year following the policy of student intake and staff and student ratio. The floor area is found to be a relatively stable independent variable since the population of students and staff is limited by the space available. The suitability of existing guideline compare with the new developed model is justified through the root mean square error.
ANALYSIS AND RESULTS
From the scatter plot of peak parking demand versus floor areas in Figure 1, regression analysis is used as a curve fitting to yield the relationship of parking demand and floor area. The result of linear relationship gives the following equations:
--- 1
Where
- Estimated peak parking demand
- Floor Area
From the analysis, the adjusted R2 value for the model is 0.72. The coefficient of the independent variables (floor area) and intercept are significant at the 5% level from the results of t-Statistic and p-value. It shows that the coefficient and intercept are acceptable. In this study the confidence interval for the coefficients of the independent variables and intercept are determined with a 95% degree of confidence.
Figure 1 : Plot of Peak Parking Demand (PPD) Versus Floor Area (FA)
VALIDITY CHECKING WITH EXISING GUIDELINE
By comparing the estimated parking demand and observed parking demand in Figure 2, it is obvious that the predicted demand from the developed model is lower than the predicted demand in the guideline (MBSA, 1995). The differences of estimation vary from a minimum of 40 parking spaces to a maximum value of 318. The percentage of additional parking demand predicted from the guideline over the predicted parking demand from the developed model differs 13% to 117%. For a quantitative comparison, root mean square error (RMSE) is used to determine the reliability of estimation.
--- 2
Where
- Predicted Peak Parking Demand from the developed Model or Guideline.
- Observed Peak Parking Demand from the Cordon Survey
- Number of samples
The RMSE of the developed model is given as 48.2 which is slightly smaller than the RMSE of the guideline as 49.4. It shows that the model gives a better prediction compared to the guideline.
Figure 2 : Comparison of Peak Parking Estimation Between Guideline and Development Model
CONCLUSION
Basically, the parking supply in campus was designed based on the guideline of the local council (MBSA, 1995). By comparing the predicted peak parking demand of the developed model with the existing guideline, it shows that the parking demand based on the developed model is different by 13% to 117% than the parking demand designated by the guideline. But the former gives more reliable estimates than the latter by looking at its RMSE. Thus, the model developed in this study gives a slightly better estimate of the parking demand with floor area as the independent variables in determining the requirement of parking provision. Furthermore, building floor area could be easily estimated prior to the design and construction stage compare to the total population of occupants later.
However, mathematical models have their limitation such as the parameters considered and/or assumption made. The numbers of study areas, also, may significantly influence the accuracy of a model. The more the study areas allow wider aspect of observation and therefore give more accurate result. In parking study of campus, the location of campus, for example in urban, sub-urban or city-campus, is also critical where it directly affect the vehicle ownership in campus. In Malaysia, population in urban area shows higher possibility of owning a vehicle. Thus, campus situated in urban area may face greater vehicle ownership if no specific policy or regulation established to control it. However, if any necessary policy or regulation is being set, it would again adjust the parking demand in campus, and thus model needs to be recalibrating for changes to suit the new condition.
Recommendations of Solution
In long-term solution of parking facilities, parking demand model assists in estimating demand hence guide the appropriate supply. During the planning of new development of faculties, this model could be used as guidance in prediction of parking supply. In long run also, the university should recognize future structured parking as an integral component of further campus development. For instance, allow parking garages in the building either as the basement or on roof to save up land. In short-term solution for those existing parking facilities, problem of inadequacy of parking facilities could be eased through either by control of demand, or considered adjustment of current supply condition. For instance, encouraging people to commute by some mode other than driving is a possible solution. The others involve an increase in the efficiency use of the current parking supply, modify the parking lot design and formulate new policy or establish new regulation.