**3. Analyzing Variability in CO2 Emissions**

#### *3.1. Overview of Activity Data Collection*

In order to estimate CO2 emissions, it is necessary to estimate the amount of oil consumption by measuring construction equipment operation time and equipment specification information. Therefore, data were collected on equipment operation time at the activity level (Table 1).


**Table 1.** Overview of the construction site with activity data collected.

This study collected data about concrete work for a typical floor slab. In this study, it was difficult to collect activity data for all tasks. Therefore, this study collected data focused on heavy equipment, which uses heavy oil in concrete pouring work; we installed a camcorder and a stopwatch to measure the operating time of each piece of construction equipment, namely, concrete pump cars and concrete mixer trucks.

Based on the specifications of the construction equipment, this study surveyed the equipment that meets the applicable standards of construction machinery expenditure as illustrated by Korean professional construction association information.

#### *3.2. Calculating Emissions from Equipment Operation Times*

In this study, one work cycle of concrete pouring work was defined from when a concrete mixer truck enters to when the truck leaves the construction site. A total of 37 work cycles were measured. However, interrupted working hours during the work cycle were excluded from the measurement.

The equipment operation time for each detailed task was measured, and 21 data sets were acquired in five categories and used in the analysis. The work of a concrete mixer truck was classified into four detailed tasks: Stopping after entering, waiting after stopping, pouring concrete, and leaving. The detailed work of the concrete pump cars was classified as a single task: Pumping concrete.

To estimate the working time of a concrete pump car for one concrete mixer truck (6 m3), we referred to the average productivity value of the amount pumped per hour (87.5 m3/h) provided in the Construction Equipment Cost Estimates Table (CECET). The CECET is an annual national standard released by the Construction Association of Korea (CAK) and the Ministry of Land, Infrastructure and Transport (MOLIT) and is calculated and published using data on the hourly hire of machines, miscellaneous materials, and drivers of construction machines listed in the standard production unit system. In this study, the concrete mixer trucks and concrete pump cars used at the construction site were collected through the CECET.

In this study, the CO2 emission coefficients of materials and equipment were used according to the data presented in the 2006 IPCC guidelines [23]. The IPCC provides CO2 emission coefficients through its own data surveys and reports submitted by each country. Recently, each country has used IPCC guidelines to calculate CO2 emission coefficients that fit the situation for that country [17].

This study does not consider the working time of concrete mixer trucks at the work planning phase because the number of concrete mixer trucks is calculated at the time of work planning. The working time of the concrete pump car was estimated using the formula (Equation (1)) proposed in the standard for reinforced concrete construction estimation [24]:

$$T\_p = t\_p \times f\_1 \times f\_2 \times Q \tag{1}$$

where:

*Tp*: Concrete pump car working time;

*tp*: Reference time, estimated time it takes for the concrete pump car to pour 1 m2;

*f*1: Facility type consideration coefficient;

*f*2: Concrete mixer truck entry condition coefficient; and

*Q*: Workload.

In this study, a reference time (*tp*) value was applied—1.25 min corresponding to 15 cm of the reinforced concrete slump—by referring to the information provided on the site where the activity data were collected [24]. The working time was calculated under the conditions that *f* <sup>1</sup> is normal (1.2) and *f* <sup>2</sup> is good (1.0) [24].

The average and the standard deviation of the measured equipment operation time are shown in Table 2. The planned equipment operation time is expressed as a single value. The variation of work time was the largest for waiting after stopping and was relatively low for stopping after entering and leaving. The average measured operation time of concrete pumping work was calculated to about 50 s longer than the value given by the machine cost calculation table. In addition, the concrete pumping work was more than twice the average measurement time in the case of the bills of quantity.


**Table 2.** Measured operation time and planned operation time (in seconds).

## *3.3. Estimating CO2 Emissions*

In this study, CO2 emissions from the oil consumption of construction equipment were estimated. The oil consumption was calculated using the measured working time at the construction site, the fuel efficiency of the construction equipment, and the engine load factor of construction equipment (Equation (2)):

$$F = T \times FE \times LF \tag{2}$$

where:

*F*: Oil Consumption by Work;

*T*: Working time;

*FE*: Fuel efficiency of construction equipment;

*LF*: Engine load factor of construction equipment.

The fuel consumption of the construction equipment is based on the equipment specification information (Table 3). The engine load factors are: Low (20–30% output, short distance travel); medium (30–40% output); waiting after stopping (when output is idling at 10%); and acceleration (100% output and continuous acceleration) [24]. When output is 40–50%, it indicates that the slope has suddenly changed, or there was a long distance of high driving resistance.


**Table 3.** Engine load factor by detailed operation.

Using the estimated amount of oil consumption, CO2 emissions for detailed tasks were calculated using Equation (3):

$$CO2 = \text{ } F \times TE \times CF \times \frac{44}{12} \tag{3}$$

where:

*TE*: Petroleum conversion coefficient; and

*CF*: Carbon emission coefficient.

In this formula, the type of oil consumed is converted to equivalent oil consumed using an oil conversion factor, since the amount of CO2 consumed differs depending on the type of oil. The converted value is multiplied by the CO2 emission factor to estimate the CO2 emission. The carbon molecular weight ratio to the carbon atom is multiplied by the constant 44/12 to calculate the final CO2 emitted. In this study, the diesel oil conversion factor is 0.000845 toe/ton, and the emission factor is 0.837 ton-C/toe (where toe is ton of oil equivalent).

#### *3.4. Comparative Analysis of Emissions*

The average CO2 emissions calculated based on the measured operating time and the planned operating time of the equipment were compared. Concrete mixer trucks do not have planned emissions, so they cannot be compared with actual CO2 emissions. A descriptive analysis of the CO2 emissions for the activities and the whole cycle was conducted to examine the variability of CO2 emissions data (Table 4).

**Table 4.** Descriptive statistics of CO2 emissions for measured operation time (gCO2).


The standard deviation, similar to the mean value, appeared to have a large value in the order of pumping concrete, pouring concrete, waiting after stopping, stopping after entering, and leaving. However, the coefficient of variation had the largest value of 1.3 for waiting after stopping, followed by stopping after entering with 0.8, leaving with 0.4, and pouring and pumping with 0.3. Therefore,

the variation was relatively small in pouring concrete, pumping concrete, and leaving, while it was relatively large for waiting after stopping and stopping after entering. On the other hand, the coefficient of variation for the whole cycle was similar to that of pouring concrete and pumping concrete. Similarly, this is the reason why the emissions of both activities have a high emission rate.

Overall, the amount of CO2 emissions during pouring works is influenced by the high emission ratios of pouring and pumping. Thus, the distribution of emissions and the total emissions from these two activities represent a relatively regular form of distribution. However, since the coefficient of variation exceeds 0.3 for both the detailed and the complete work, the analyzed deviation is relatively large. Moreover, several activities showed a relatively high coefficient of variation and a biased distribution in the emission data even though the emission rate was low. Thus, the estimation of CO2 emissions using a single value may cause a significant error due to an insufficient reflection of the variability. In order to increase the reliability of CO2 emission estimation, it is necessary to consider the variability of CO2 emissions for each activity.
