Page 70 - Contributed Paper Session (CPS) - Volume 3

CPS1946 Jittima D. et al.
Center (REIC), an agent taking charge of recording number of condominium
units launched. Findings shown in table below.

Table 1. The number of new residential units and the number of electric meters

Data 2013 2014 2015 2016 2017 Total
New residential 67,267 65,313 53,626 56,721 57,563 300,490
units
Electric meters 48,616 45,679 56,659 62,324 47,849 261,127
(%) (72.3) (69.9) (105.7) (109.9) (83.1) (86.9)
Source: MEA, REIC and calculated by BOT
Remarks: Two data sources for new residential units launched in 3 territories
from 2013-17.

Results are concluded as follows:
1. The 5-year average ratio between registered electricity meter and REIC
data stands around 87%. This exhibits meter data can be treated as one
proxy for estimating new residential units launched in each period.
2. In some years, numbers of meter data exceed new residential units
launched, for instance, in 2015-16. This was caused by cumulative effect,
for new unsold units launched in 2013–14 were sold-out, also had meter
registered in years after.

To model the Occupancy Rate (OR), one presumption of un-occupancy
units is required. In the study, use of 20 kilowatt-hours (kWh) or above energy
3
consumption per month is set as a “Threshold” for occupied units. For other
group, units with power utilization less than 20 kilowatt-hours shall be treated
as unoccupied. It follows naturally that occupancy rate can be calculated by
taking number of units with electricity usage equal or above the Threshold,
dividing by total number of units (Equation 1). If OR is high, this means
demand for residential units is elevated. The relationship also holds in the
opposite direction.

Number of occupied units
Occupancy Rate (OR) = (Equation 1)
Total number of units

3 In the study, a Threshold of electricity usage (in units of kilowatt-hours per month) is imposed,
but not non-occupancy duration criteria as defined by Ecotagious (2016). To improve the
Threshold’s validity and precision, non–occupancy period such as students’ non-occupancy
during school holidays should be factored-in to calculate average energy consumption.
Following this concerns, the NOR and OR can reflect better reality. On balance, the computed
OR without imposition of duration criteria can be regarded as minimum occupancy level, actual
residency rate could be more.
59 | I S I W S C 2 0 1 9