Journal of Cleaner Production 463 (2024) 142646

Contents lists available at ScienceDirect

Journal of Cleaner Production

journal homepage: www.elsevier.com/locate/jclepro

2030 oil palm plantation carbon footprint estimation using O-LCA

and forecasting
Farizal F a, c, *, Trisha Amanda a, Muhammad Dachyar a, Zainura Zainon Noor b, d
a
Department of Industrial Engineering, Faculty of Engineering, Universitas Indonesia, Kampus Baru UI Depok, 16424, Indonesia
b
School of Chemical and Energy Engineering, Faculty of Engineering, Universiti Teknologi Malaysia, Johor Bahru, 81310, Malaysia
c
Center for Biomass Valorisation (RCBV), Universitas Indonesia, Indonesia
d
Centre for Environmental Sustainability and Water Security (IPASA), Universiti Teknologi Malaysia, Malaysia

A R T I C L E I N F O A B S T R A C T

Handling Editor: Kathleen Aviso Since palm oil products have a crucial role in society development, palm oil business in Indonesia has grown
significantly over the past few decades. However, this industry is frequently linked to environmental issues,
Keywords: particularly its potential emission of greenhouse gases (GHGs). This study attempts to estimate the carbon
Palm oil footprint of a palm oil company’s operation for the year of 2030. In doing so, a novel methodology that consists
O-LCA
of organizational lifecycle assessment (O-LCA), simple linear regression (SLR), and double exponential
Forecasting
smoothing (DES) methods is proposed. O-LCA is used to identify the sources of emission and estimate the amount
Carbon footprint
Greenhouse gases of the emission generated, while SLR and DES are used to forecast the sources. As the result, the carbon footprint
in 2030 is estimated to be 62,758,433.56 kg Carbon Dioxide equivalent (CO2eq) where on average a ton of crude
palm oil (CPO) produces 1.08 ton CO2eq. The study also discloses that the three largest emission sources are palm
oil mill effluent (POME), fertilizer, and transportation. The forecasting methods used are quite accurate with
mean absolute percentage error (MAPE) of less than 10%. The results of this study can shed a light to help
Indonesia achieving its target to reduce GHG emission by 2030.

1. Introduction number one producer in the following years. Its palm oil plantation has
significantly expanded during the past few years. Fig. 1 shows the oil
Oil palm (Elaeis Guineensis Jacq) is one of the most efficient vegetable palm plantation area from 2017 to 2022 (Kementan, 2020). Between
oil producers in terms of quality per unit area, uniformity of production, 2017 and 2021, the plantation area grows steadily on average by1.8%
and by-products. Its utilization covers wide areas of human need. Oil per year. The growth rate is even higher from 2021 to 2022, which is
palm flesh produces crude palm oil (CPO) that can be processed into 8.6%. In Indonesia, besides rubber and coffee, palm oil is the largest
food ingredients such as cooking oil and margarine, as well as the non- source of foreign exchange earnings from the non-oil and gas sector.
food ingredients (oil chemicals) such as candles, soap, cosmetic in­ Besides the need for export, there is a concern in palm oil utilization due
gredients, and diesel fuel. Furthermore, palm oil fruit juice fibre can be to the ever-increasing domestic need (Khatiwada et al., 2021). The need
used as feedstock to produce board, pulp, energy source, and animal goes up due to the increased Indonesian population and higher utiliza­
feed (Yilmaz and Ağagündüz, 2022). To date, the global palm oil usage tion in few sectors. On the other hand, environmental concerns, where
increases significantly. From 2015 to 2022 palm oil consumption growth CPO production is deemed as a source of carbon release, putting another
is 3.68% on average, with decline only in the year of 2021–2022, when obstacle for Indonesia to exports CPO to the European and American
the lockdown due to COVID-19 pandemic struct the world. However, the nations (Uning et al., 2020; Rum et al., 2022).
growth consumption rate rebounded almost double on the following In Indonesia agriculture contributes 13% of total emissions. It was
year, i.e. by 6.92% (Shahbandeh, 2023). estimated that the emissions in this sector in the year of 2030 is
Among palm oil producers, Indonesia is ranked the first. In 2018, it 478,503.66 kg ton CO2eq (Bappenas-LCDI, 2024). As many countries in
produced 41 million metric ton of palm oil which accounted for 57% of the world, Indonesia has pledged to reduce its future GHG emissions.
the world supply (Ritchie and Roser, 2021). Indonesia is still ranked the Indonesian government has renewed its GHG emission reduction target

* Corresponding author. Department of Industrial Engineering, Faculty of Engineering, Universitas Indonesia, Kampus Baru UI Depok, 16424, Indonesia.
E-mail address: [email protected] (F. F).

https://doi.org/10.1016/j.jclepro.2024.142646
Received 19 November 2023; Received in revised form 10 May 2024; Accepted 20 May 2024
Available online 21 May 2024
0959-6526/© 2024 Elsevier Ltd. All rights are reserved, including those for text and data mining, AI training, and similar technologies.
F. F et al. Journal of Cleaner Production 463 (2024) 142646

feedstock for making biodiesel than microalgae since it has larger energy
ratio and produces lower GHG emissions. However, considering the
feedstock availability in the long run, microalgae is a better source since
it is not conflicting with the food sources.
Numerous LCA studies have been conducted on palm oil. For
instance, Sawangkeaw et al. (2012) compared the energy used to pro­
duce N and C supercritical alcohol from crude palm oil, while Queiroz
et al. (2012) used LCA to assess biodiesel production. Rivera-Méndez
et al. (2017) calculated carbon footprint from processing fresh fruit
bunches in the palm oil plantation. Lam et al. (2019) studied the
greenhouse gas footprint due to palm oil plantation expansion by 2030.
Hashim et al. (2018) estimated the carbon footprint of palm oil planted
on peat soil. A more recent study, Su et al. (2023) used different tech­
nology, which is pyrolysis to produce biofuel while Pranolo et al. (2023)
used gasification process to produce fuel gas. Instead of estimating
carbon footprint of palm oil agri-culture, Sabli et al. (2017) developed a
Fig. 1. Palm oil plantation area over the years from 2017 to 2022 (Kementan, methodology to estimate water footprint of the industry. CPO utilization
2020) is quite vast, which includes the making of cosmetics, food, material
*predicted data. mix, etc. Martinez et al. (2017) investigated the environmental impact of
using refined palm kernel oil to make cosmetic cream, while Francke
by 29% by 2030 (PP., 2021). This study estimates palm oil carbon and Castro (2013) analysed carbon and water footprint of making soap
footprint at an oil palm plantation at the Province of Jambi in the year bar. Hamada et al. (2020) reviewed papers that utilize palm oil clinker
2030. For the purpose, the study combines O-LCA and Double Expo­ as concrete aggregate material. Studies on CPO as one of the food in­
nential Smoothing (DES) and/or Simple Linear Regression (SLR) fore­ gredients in cereal and biscuits were conducted by Jeswani et al. (2018)
casting methods. The forecasting methods are used to estimate and Noya et al. (2018), respectively.
inventory in the year of 2030 while O-LCA is used to estimate the carbon Selecting a technology that has the least negative impact to the
footprint. Nonetheless, the scope of the study is only activities under the environment is sometimes inadequate, so LCA is combined with another
control of the palm oil company under consideration. However, the method to ensure the selected technology is more viable. Vazifeh et al.
study does not only estimate the existing carbon footprint inventory, but (2023) combined LCA with techno-economic analysis to evaluate the
it expands the findings by forecasting the footprint for 2030. As palm oil feasibility of biomass conversion technologies in producing lumber, in­
is one of the major agriculture products in Indonesia, by doing so, the dustrial sugar, and wood pellets. LCA does not consider spatial and
findings in this study can open a way to address environmental concerns dynamic of the system when assessing the environmental impacts of a
of the sector as well can be used as a basis to select alternative way(s) to product or process. This may lead to misinterpretation of the impacts.
reach the target. McAvoy et al. (2021) reported many LCA studies that combined with the
system thinking approach to address the concern. To ensure the data
2. Literature study on LCA used on LCA is more tractable, LCA is combined with blockchain tech­
nology (Sou and Domenech, 2022). A case study on leather handbags
One popular way to study environmental impact is using life cycle showed that the technology helps to make the data used tractable and
analysis (LCA). LCA is a tool to analyse the environmental impact of a reliable in identifying the environmental hotspot and circular potential.
product, process, or service during all stages of the product life cycle. A technology may have small impact on the environment, but it may
The assessment covers from the cradle to the grave of the product (US incur high cost during the termination phase. To implement LCA and
EPA, 2006). Since its inception in 1990s, there are many LCA studies. economic evaluation, Gennitsaris et al. (2023) combined LCA with data
For instance, the ones that deal with new and renewable energy (Kumar envelopment analysis (DEA). They used the methods to evaluate wind
and Pal, 2023; Lee et al., 2023; Farizal & Septia, 2015), alternative ve­ turbine decommissioning in Greece. In building environmental impact
hicles (Oda et al., 2022; Pereira and Coelho, 2015), new materials assessment, Tam et al. (2023) combined LCA with building information
(Vieira et al., 2023; Hesser, 2015), consumer plastics (Deeney et al., modeling (BIM). Integrating LCA with analytic hierarchy process (AHP)
2023), and food products (Casolani et al., 2022; El Joumri et al., 2023). was conducted by Wei et al. (2022). They used the methods to evaluate
Solar panel well believed causes zero emission when it is operated. the benefits of replacing disposable plastic tableware in airline catering
However, an LCA study conducted by Farizal & Septia (2015) showed with the ones made of stainless steel, bamboo or bio-based plastic. AHP
that the technology is not that environmentally friendly. Solar panel is used to determine the replacement alternatives that fit the economic,
production involves the consumption of pure silicon extracted from safety, and aesthetic factors.
silicon dioxide (SiO2) which is mainly found from sand on earth. Sand In 2015 the United Nations Environment Program (UNEP) and So­
mining to produce solar panel spoils seashore ecosystem including ciety of Environmental Toxicology and Chemistry (SETAC) developed a
freshwater aquatic ecotoxicity, marine aquatic ecotoxicity, and water new approach known as the Organizational Life Cycle Assessment (O-
eutrophication. These environmental impacts are worsened for the solar LCA). Like LCA, O-LCA is also a tool to evaluate environmental impact.
panel equipped with battery (off-grid system). Furthermore, silicon The difference between LCA and O-LCA is in the terms of the assessment
extraction and solar panel production consume energy that mostly scope. LCA is usually focused on life cycle assessment of a product
supplied from coal-fired power plant. The plant notoriously produces whereas O-LCA is focused on the company’s annual activities. The
GHG emission. Lee et al. (2023) studied the impact of utilizing scopes under O-LCA include both direct and indirect value chain oper­
non-recyclable municipal solid waste as the feedstock to produce liquid ations (Martínez-Blanco et al., 2015). Since every organization has
fuel using gasification and Fischer-Tropsch conversion technology. Even unique characteristics, O-LCA may be used in many ways. O-LCA ana­
though this work will address waste issue and energy problem concur­ lyses the entire organization, not only company facilities but also the
rently, Lee et al. found that its carbon intensity (emission) is higher than upstream and downstream activities (Martinez-Blanco et al., 2015).
that of petroleum fuels. Kumar and Pal (2023) used LCA to compare O-LCA enables businesses to identify critical environmental operations
energy ratio and GHG emissions of producing biodiesel from freshwater and make improvement by considering a variety of sustainability vari­
microalgae and Karanja. The results showed that Karanja is a better ables as well as economic, social, and technological elements.

2
F. F et al. Journal of Cleaner Production 463 (2024) 142646

Since its inception, unlike LCA, the studies that adopt O-LCA to assess of Goal and Scope, Life Cycle Inventory Analysis, Life Cycle Impact
how business operations affect the environment are limited (Rimano Assessment, and Interpretation. Fig. 2 illustrates the overall methodol­
et al., 2019). To date, there are only 12 papers published in journal, so ogy used in this study which is the integration of O-LCA and double
on average it is less than 3 papers each year. However, despite its lack of exponential smoothing (DES) and simple linear regression (SLR) to
utilization, scientific community recognize the importance of the new forecast the carbon emission of a palm oil company for the year of 2030.
LCA method demonstrated by the finding of Martinez-Blanco et al. As in LCA, definition of objectives and scope is the first phase of O-
(2018) who conducted road testing on O-LCA worldwide. One of the LCA. The objectives must be clearly defined as they are crucial for all the
earliest publications on O-LCA is Manzardo et al. (2015). They used the remaining O-LCA phases. Furthermore, in determining the research
method in the beverage packaging sector. Resta et al. (2016) utilized objectives, ISO/TS 14072 needs to state that the results are not to be
O-LCA to improve environmental management in the textile sector. On used in comparative statements that are aimed to be publicly disclosed
the same year, Jungbluth et al. (2016) studied the impact of food con­ (Martínez-Blanco et al., 2015). In the O-LCA study, the scope drawn
sumption in canteens to the environment, however, they used LCA must be specific and well-defined to ensure that the study’s breadth,
instead of O-LCA. Rimano et al. (2021) studied the environmental depth, and details are compatible and sufficient to achieve the stated
impact of organizations in packaging industry. Cucchi et al. (2022) used objectives (ISO 14040, 2006). Practitioners should be sure to describe
O-LCA in the study of a ceramic tile manufacturer in the industry 4.0 the organization under study, the constraints placed on the organiza­
setting. A recent O-LCA study at a corporate group that generates elec­ tion’s life cycle (which ideally equates to the boundary between eco­
tricity from renewable and nonrenewable sources was conducted by nomics and nature), where the data shall come from and of the required
Notarnicola et al. (2023). A paper that combines O-LCA with product life quality, and how the information will be obtained. As defined by ISO
cycle method is published by Toniolo et al. (2023) to discover envi­ 14040 (2006), the system boundary in a product of LCA is a set of
ronmental benefits in recovering steel byproducts. criteria that determine which unit processes are part of the production
system and thus determine which processes should be included in LCA
3. Research methodology (Klöpffer and Curran, 2017).
Life Cycle Inventory (LCI) involves the stages of data collection from
Following the guideline published by UNEP and SETAC, the O-LCA the company’s life cycle. The objective is to measure the relevant inputs
addresses analytical, managerial, and social objectives. Applying this and outputs from the company’s activities, for example energy and
method in a company involves a better understanding of environmental feedstock requirements, atmospheric emission, water-borne emission,
performance in the life cycle, thus involving impacts that are directly solid waste, and other discharges for the entire system life cycle
controlled by the company and emission due to upstream and/or (Hauschild et al., 2018). The O-LCA Guidelines provide a framework in
downstream activities that impact consumers. This step will influence categorizing activities into direct upstream activities, indirect upstream
environmental management and strategic decisions (Sparrevik and activities and indirect downstream activities (Forin et al., 2019).
Utstøl, 2019). The O-LCA approach consists of four stages: the Definition

Fig. 2. Methodology diagram of the study.

3
F. F et al. Journal of Cleaner Production 463 (2024) 142646

1. Direct activities: Activities that are directly owned or controlled by fungicide, fertilized with NPK, and equipped with water irrigation.
the organization under study including, for the producing company, Planting oil palm tree does not only occur when opening a new plan­
all basic flows associated with the organization’s facilities. tation but also when rejuvenating an old/unproductive one.
2. Indirect activities (portions which are not part of the reporting or­ Palm oil fresh fruit bunches are harvested and loaded to the trucks
ganization) are classified as: that transport the fresh fruit bunches (FFB) to the processing plant. In
a. Indirect upstream activities: if the activities are carried out by the plant site, they are sorted manually according to their quality and
upstream suppliers or supporting organizations (e.g., purchasing then processed through some stations (sterilization, threshing, pressing,
raw materials and semi-finished products, outsourcing services separation, and clarification) to get CPO and kernel.
such as cleaning, IT or marketing, external waste treatment,
capital equipment such as machinery and buildings, transporting
materials to the company); 3.1. Carbon footprint quantification
b. Indirect downstream activities are performed once the product
leaves the firm (e.g., transport of the product from the organiza­ Using life cycle inventory findings, LCI assessment aims to evaluate
tion to the customers, logistics, usage phase, end-of-life phase). the significance of potential environmental impacts. The process links
inventory data to certain environmental impact categories, in this case
In this study, all palm oil related activities are projected to future the greenhouse gas emission. This stage also includes the interpretation
activities, i.e. activities in the year of 2030. Based on historical data of the impact. In this study, the impact category is assessed according to
collected from 2012 to 2021, the projected values are forecasted using the carbon footprint (Luisetti et al., 2020). The carbon footprint meth­
DES or SLR. Selection of the forecasting methods (either DES or SLR) for odology is defined as the quantification of greenhouse gas emissions
each activity depends on the nature of the data (Farnum and Stanton, associated with the life cycle of goods or services. Carbon footprint CF is
1989). Calculating forecasting error is the way to check the goodness of calculated using Equation (1). CF is presented in the unit of Carbon
the model. Dioxide Equivalent (kg CO2eq).
Fig. 3 shows activities and flows to produce CPO from oil palm. From Carbon Footprint (CF) = A x EF x GWP (1)
the site where the activities taken place, palm oil production process in
broad view can be split into two parts, which are plantation site-related Where A is the activity data, EF is the emission factor and GWP is the
activities and processing plant site-related activities. In plantation site, global warming potential weight.
oil palm tree is cultivated from its seed. Seeding is conducted at a According to the Intergovernmental Panel on Climate Change
nursery camp. Once the plant reaches a certain size, it is moved to the (IPCC), carbon footprint is generated from energy-related, industrial,
plantation area to be planted, nurtured, and harvested. To make sure agricultural sectors, and from waste. Carbon footprint of the stationary
that the oil palm grows healthy, the tree is sprayed with insecticide and combustion emission CFSC is estimated using Equation (2), while for
non-stationary combustion, it is estimated using Equation (3).

Fig. 3. Activities to produce CPO and kernel from oil palm.

4
F. F et al. Journal of Cleaner Production 463 (2024) 142646

( )
emission factor of gas pollutants for vehicle category j with technology k
CFSC = FC x LHV x EFFC,Technology x Mr GHG/ x EFPG (2)
Mr PG (in kg CO2/km), rH:c is the ratio of mass of hydrogen to carbon, ro:c is the
ratio of mass of oxygen to carbon, FCCALC k,m is the fuel consumption of
Where FC is the fuel consumption of the combustion engine (in kg), LHV vehicle j with technology k (in kg fuel).
is the calorific value of the fuel (in TJ/kg), EFFC,Technology is the emission [( ) ]
factor from combustion that uses a certain technology (in kg GHG/TJ), CF WW = TOWj − S(j) + EF(j) − R(j) (10)
Mr GHG is the relative mass of the greenhouse gases, Mr PG is the
Relative Mass of the precursor gas, and EFPG is the emission factor of the Where TOWj a is the total materials that can be organically degraded in
precursor gas. liquid waste from industry i in the inventory year (in kg COD/m3), S(j) is
( ) the organic component that is disposed of in the form of sludge in the
CFOR− M = FC x LHV x EFFC,Technology x ρFC x Mr GHG/ x EFPG (3) inventory year (in kg COD/m3), EF(j) is the emission factor for industry i,
Mr PG
and R(j) is the amount of methane recovered to form the energy (in kg
CH4).
Where ρFC is the fuel density (in kg/liter).
The formula to calculate carbon footprint of fertilizer, dolomite, CFPesticide = Active ingredients × FEP (11)
transportation, POME treatment, and pesticide are demonstrated by
Equations ((4), (5), (7), (10) and (11), respectively. Where Active ingredients is the ingredients in pesticide that are biologi­
( ) cally active (in gram/liter), FEP is the Pesticide Emission Factor (in kg
CFFertilizer = FSN × FracGASF × Mr GHG/ × EF4 GHG/liter).
Mr PG
( ) (4) Interpretation is the last phase of LCA where results from the pre­
+( FSN ×FracLeach × EF5 ) + Consumptionfertilizer × EFGHG,Fertilizer vious phases are consolidated. The results are also analysed in terms of
uncertainty of data obtained and the assumptions that have been made
Where FSN is the amount of N fertilizer applied to the soil, FracGASF is the and documented during the study. This stage is useful to explain in detail
standard value of vaporized nitrogen in the form of NH3 and NO2, which and how specific elements of the inventory contribute to envi­
FracLeach is the standard value of nitrogen that released or lost, FE4 is the ronmental impact, which impact categories are the most relevant, or the
Emission Factor of volatilized nitrogen in the form of NH3 and NO2, EF5 life cycle phase with the greatest impact (Hauschild et al., 2018).
is the Emission Factor of the nitrogen released, and EFGHG is the Upon calculating the impact indicators of the life cycle, the results
Greenhouse Gas Emission Factor in the manufacturing process. should be validated, and at a minimum, this should include a reality
assessment check of the expected results to identify the data or modeling
CF = Consumptiondolomite x EFDolomite (5)
Dolomite
errors. Once the model general integrity has been verified, the results
can be further analysed to identify the sensitivity, the main contributing
Where Consumptiondolomite is the amount of dolomite applied and
factors, and differences between the options in the improvement sce­
EFDolomite is the Dolomite Emission Factor (in kg GHG/kg Dolomite).
nario. Conclusion and recommendations from the interpretation phase
To calculate the Dolomite carbon footprint, the second part of
will support internal decision-making and establish a framework for
Equation (5) is further defined as Equation (6):
monitoring environmental performance. Recommendations should be
( )
Mr CO2 ( ) based on the research’s conclusions after improvement analysis has been
EFDolomite = × FEAplikasi,Dolomite × Massa CaMg(CO3)2
Ar C carried out to meet the research objectives. In this phase, important facts
( ) about the application of LCA must be considered regardless of which
+ FEProduksi,Dolomite × Massa CaO.MgO (6)
interpretation method is chosen. This methodology is vital in evaluating
Where Mr CO2 is the relative mass of molecule CO2, Ar C is the relative environmental and human health impacts (Kjaer et al., 2018).
mass of atom C, FEAplikasi, Dolomite is the emission factor of Dolomite
production in the form CaO.MgO, FEproduksi, Dolomite is the emission 3.2. Forecasting and error
factor of Dolomite usage in the form CaMg(CO3)2, Massa CaMg(CO3)2 is
the Dolomite mass before its reacted with the surrounding (in kg), Forecasting means predicting the future event(s) based on past and
whereas Massa CaO.MgO is the Dolomite lime mass after reacted with current data (Montgomery et al., 2008). In this study, the amount of
the surrounding (in kg). carbon footprint is estimated for the year 2030. To do so, the input
∑ variables, such as the consumption of fuel and materials (fertilizers,
CF Transport = Nj,k × Mj,k × FEi,j,k (7) herbicide, or insecticide), that described according to Equations (2)–(8)
k
are forecasted using Double Exponential Smoothing (DES) and/or Sim­
ple Linear Regression (SLR) methods. The methods are selected due to
Where FEi,j,k is the Pollutant Emission Factor i for the vehicle category j
their simplicity and easy to use. A forecasting method is used not
with technology k (in kg GHG/km), Mj,k is the average distance travelled
because of its complexity or sophisticated approach. Rather, the method
by vehicle j with technology k (in km), and Nj,k is the number of vehicles
will be used if it is able to provide quality forecasting results. Mitchell
category j with technology k.
et al. (1975) already listed 150 forecasting approaches. To determine the
To calculate the transport carbon footprint, the last part of Equation
quality, forecasting error is calculated (Montgomery et al., 2008; Far­
(7) is further defined as Equations (8) and (9). Equation (8) is the
num and Stanton, 1989; Farizal et al., 2021). For regression method,
formulation due to carbon dioxide pollutant while Equation (9) is the
Coefficient of Determination (R2 or R-sq) is also a measure to evaluate
formulation due to other pollutant gases such as methane (CH4), nitrous
the goodness of the model.
oxide (N2O), and biogenic CO2.
One popular method to measure forecasting error is the mean ab­
FCCALC solute percentage error (MAPE) (Kim and Kim, 2016). MAPE expresses
k,m
FECO2,j,k = 44.011 × (8) the error in the form of percentage. This means MAPE is dimensionless
12.011 + 1.008rH:C,m + 16.000rO:C.m
or unit-free. Besides being easy to interpret its results, MAPE is useful to
FEGRK,j,k = FEGRK × Mr GHG/ × EFPG (9) compare the accuracy of similar or even different forecasting models on
Mr PG
two distinctive series (Farnum and Stanton, 1989; Byrne, 2012). MAPE
Where FECO2,j,k is the emission factor of carbon dioxide pollutant for is calculated using Equation (12):
vehicle category j with technology k (in kg CO2/km), FEGRK,j,k is the

5
F. F et al. Journal of Cleaner Production 463 (2024) 142646

n |
yi − ŷi | deemed to have similar impact to each palm oil tree. This assumption
∑ x 100%
MAPE =
yi
(12) enables the estimation of the total substances required.
i=1
n
Based on the activities highlighted in Fig. 3 and the related data
Where yi is the actual dependent variable values, ŷi is the predicted collected from 2012 to 2021, the details of O-LCA approach integrated
dependent variable results given by the model, and n is the number of with DSL and SLR forecasting methods are as follows.
observations.
A forecasting method is considered good if its MAPE is 10–20%. If the 4.2. Palm oil organization life cycle analysis
error is less than 10% the method is considered as excellent (George
et al., 2015; Fotios et al., 2013; Gilliland, 2010). 4.2.1. Phase 1: goal and scope definition
Coefficient of Determination, R-sq is calculated using Equation (13). The first step of O-LCA is defining the goal and scope of the study.
∑
(yi − ŷi )2 The goal is to estimate the carbon footprint in palm oil agri-industry,
R2 = 1 − ∑i (13) whereas the scope encompasses all activities of the industry. Accord­
(yi − y)2
i ing to O-LCA, palm oil activities as depicted in Fig. 3 can be categorized
as upstream and downstream activities, as illustrated in Fig. 4. The
Where ȳ is the average of the actual dependent variable values. figure shows that upstream activities are divided into direct and indirect
R2 or R-sq has a value between 0 and 1. The closer the value to 1, the activities. Direct upstream activities include palm oil nursery activities,
better the predicted dependent variable ŷi fits the actual dependent seed transportation, rejuvenation, and maintenance of palm oil planta­
variable data yi, thus the better the model in giving the prediction tions carried out by the company, and FFB transportation from the
(Farnum and Stanton, 1989; Vining, 1998). plantation areas. Meanwhile, indirect upstream activities are activities
carried out by partner farmers and independent smallholders, such as
4. Results and analysis seed transportation for the partner plantation rejuvenation activities,
maintenance activities of the partner palm oil plantations, and trans­
4.1. Oil palm plantation portation of the FFB. Direct activities are the company’s main activities
which consist of processing the FFB into the main products, palm oil, and
This study was taken place at a palm oil company located at Mestong kernel seeds.
District, the Province of Jambi, Central Sumatra. The company, estab­ In processing the FFB, palm oil mills also produce palm oil waste,
lished in 2010, runs both the plantation and palm oil mill plant. The both in the form of liquid and solid waste. The way the company handles
company manages its own plantation with the area of 1817 ha and its’ its waste determines its carbon footprint.
partner land of 7960 ha. The company employs the best seeding prac­ The company’s downstream activities involve shipment of CPO and
tice, rejuvenating the palm oil plants at a special place and then rearing kernel to distributors. In general, the company’s CPO and kernel prod­
and harvesting the trees at their plantation. The transport vehicles travel ucts are exported abroad, such as China. In this study, delivery of CPO
back and forth from/to the seeding area, the plantation site, and the and kernel products is only carried out up to the delivery to the third
processing mill plant. Running the plantation, the company produced parties. At this point, the company no longer has the control of its
FFB from 2012 to 2021 on average 291,996 ton per year. To fulfil its product delivery to the next consumers.
production target, the company purchases a small amount of FFB from
independent farmers from time to time. Processing FFB in its production 4.2.2. Phase 2: life cycle inventory (LCI)
machines, the company produced on average 55,020 ton of CPO and In O-LCA, LCI analysis involves data collection and analysis of the
13,581.3 ton of kernel. As any other palm oil processing companies, the entire company’s life cycle with the objective to measure the related
company also produces some solid and liquid wastes. input and output activities. For such purpose, data in the period of
The assumptions in this study are as follows. 2012–2021 were collected from the field. The data include the energy
used, the raw materials/feedstock consumed, emission released to the
1. Congestion indeed produces more GHG emissions than smooth atmosphere as well as drifted to the water, solid waste produced, etc.
flowing traffic (Wang et al., 2023; Bernardo et al., 2021; Armah (Hauschild et al., 2018). In this step, LCI calculates the amount of carbon
et al., 2010). However, in this study it is assumed that there is no footprint produced pertaining to palm oil industry activities as defined
congestion for the vehicles traveling back and forth from/to seedling in the goal and scope of this study. In this study, carbon footprint is
facility, plantation, and milling plant. This assumption is acceptable estimated for the year of 2030 using double exponential smoothing
since there is no congestion on the countryside. Queuing indeed (DES) and simple linear regression (SLR) methods. Forecasting in this
occurs when the trucks lining up during FFB loading and unloading. study was conducted using Microsoft Excel spreadsheets.
But while in queue, most of the time the trucks’ engine are turned off,
so they do not produce GHG emissions. 1. Estimation of oil palm rejuvenation in 2030
2. As age (or mileage) of a vehicle increases it consumes more fuel to
travel the same distance (Wang et al., 2022; Greene et al., 2017; Oil palm rejuvenation is an effort to maintain and improve planta­
Caserini et al., 2013). In this study, it is assumed that the fuel tion productivity by replacing unproductive oil palm plants with the
consumed by the vehicles used on the plantation only depends on the new ones. This effort is important. Oil palm rejuvenation can be per­
distance they travel. It does not depend on the age (mileage) of the formed either gradually or thoroughly. Considering sustainability issue,
vehicles. This assumption is justified since the vehicles are well rejuvenation must be carried out environmentally friendly and involves
maintained (Danilecki et al., 2021; Wursthorn et al., 2010; Bin, no burning (Ernawati et al., 2021). Rejuvenation must be planned and
2003). carried out carefully and in detail to avoid potential losses. It is advisable
3. Throughout its productive lifetime, every palm oil tree is assumed to to do rejuvenation gradually rather than thoroughly by dividing the old
produce FFB uniformly. This assumption enables future estimation of land into few sections (Dohong et al., 2018).
FFB production based on the harvest historical data. This assumption In this study, the palm oil company has planned its rejuvenation
is justified since the company employs plantation best practice that activities in some plantation areas in 2023, 2025, 2028, and 2030. By
maintains palm trees healthy and fitness. 2030, the total plantation areas to be rejuvenated is about 248 ha. To
4. During the nursery, rejuvenation, and rearing phases, the same estimate the carbon footprint of this activity, the data regarding con­
quantity of substances (fertilizers, insecticides, fungicides, etc.) are sumption of dolomite and diesel for each plantation, the area, and the

6
F. F et al. Journal of Cleaner Production 463 (2024) 142646

Fig. 4. Chart of up-stream and down-stream palm oil industry activities.

time when the rejuvenation takes place were collected and then fore­ appropriateness. Fig. 6 shows an example of the testing results for
casted using the SLR method. Diesel is used as the excavator’s fuel while dolomite. The results ensure the appropriateness of the SLR method. The
dolomite is used to improve the soil acidity level. The quantity of Normal Probability Plot of the model (top left figure) resembles a linear
dolomite applied to the land is 400–500 kg per hectare. This amount is line. This indicates that the data fits the normal distribution. This result
quite huge. But for rejuvenation, it is only applied once. The forecasting is supported by the Histogram chart (bottom left) that shows a very
was calculated using dolomite and diesel fuel as the response variables much normal distribution graph. The Residuals vs Fitted Values chart
and land area as the predictor. The results show that the SLR models (top right) shows randomly distributed dots with no indication of spe­
have MAPE of 2.19 and 2.39 for excavator fuel and dolomite, respec­ cific patterns such as double bowls or curvature. The chart indicates that
tively, and R-sq of 0.952 and 0.955 for excavator fuel and dolomite, the residuals have constant variances and fulfil homoscedastic
respectively. The values of MAPE and R-sq indicate that excellent fore­ assumption. Independency assumption is fulfilled from the Versus Order
casting has been conducted since the errors are very small and the co­ chart (bottom right) that shows no indication of correlation between the
efficients of determination R-sq are closed to 1. Fig. 5 shows the plot of residuals.
the actual and forecasted data for excavator fuel and dolomite. Using the
model, the predicted amount of diesel fuel and dolomite by the year of 2. Estimation of oil palm nursery (seedling) activity in 2030
2030 is 4605 L and 119,866 kg, respectively.
One requirement to use SLR method is normality and independency The number of palm oil seeds by 2030 is forecasted using the double
assumption (Vining, 1998). Normality-related tests and independency exponential smoothing (DES) method. Based on the actual number of
are conducted every time the method used to ensure the method palm oil seeds in 2012–2021, the number of palm oil seeds within
2022–2030 is forecasted. MAPE of the forecast is 1.3% which indicates
an excellent forecast. Based on the forecast, the number of seeds
required in 2030 is 65,207.
There are two types of nursery (seedling) systems, namely two-stage
nursery and one-stage nursery. Two-stage nursery involves planting
palm seed in polybags and transferring the polybags to larger polybags
at the main nursery. This system gives better palm oil plant development
than the one-stage nursery (Izzah et al., 2020). Nursery activities involve
the consumption of diesel fuel to run irrigation machines as well as apply
fertilizers, fungicides, and insecticides. Seedlings are watered with clean
water twice a day, unless it is raining for more than 7–8 mm on that day.
The amount of water showered to the polybags depends on the age of the
seeds (Syahminar et al., 2020). Watering must be conducted using a fine
spray to avoid damaging the seeds in polybags, so that the soil where the
seeds are planted will have the desired moisture and are not dense
(Appavu et al., 2019). In addition to watering the palm seeds, the trees
should be protected from fungus and insects. 2 g of fungicides are dis­
solved on 1 L of water to spay 100 seeds. For insecticide, it takes 3 g in 1
L of water. Data for the amount of fertilizers, fungicides, insecticides,
Fig. 5. Consumed dolomite and diesel excavator fuel forecasting results in the and diesel consumed in 2012–2021 were collected and used to forecast
rejuvenation.

7
F. F et al. Journal of Cleaner Production 463 (2024) 142646

Fig. 6. Findings of residual tests for dolomite.

the consumptions of the substances in 2030. fertilizing and weed clearing. NPK fertilizer that contains Nitrogen,
The quantity of fertilizers, fungicides, insecticides, and diesel fuel Phosphorus, and Kalium is used to enrich the soil nutrient so that the
required in nurturing the palm oil seeds is forecasted using the SLR trees can grow healthy. Kieserite, a fertilizer that contains Magnesium is
method. Fertilizers, fungicides, insecticides, and diesel fuel are the also needed. Meanwhile, herbicides are used to protect the trees from
response variables, while the number of seedlings is the independent the threat of fungus and insects. A palm oil tree on average consumes
variable. MAPE and R-sq of the SLR models are summarized in Table 1. 5–9 kg of NPK fertilizer. The fertilizer is given twice a year.
Table 1 shows that the accuracy of the forecasts is excellent with very To forecast the consumption of NPK fertilizer, kieserite, and herbi­
small MAPE and R-sq close to one. Fig. 7 (a) shows the plot of fertilizer cides in 2030, the data of fertilizer, kieserite, and herbicides applied to
consumed and diesel fuel used and their predicted values, while Fig. 7 the trees in 2012–2021 were collected. The amount of fertilizer and
(b) shows the plot of fungicides and insecticides consumed. Fig. 7 (a) and kieserite used by the company fluctuate, but the amount of herbicides
(b) display the accuracy of the forecasts visually. decreases over the last few years. Using DES method, it was found that
MAPE of the forecast is 0.92%, 0.42%, and 3.5% for NPK fertilizer,
3. Estimation of travel distance to transport oil palm seeds by 2030 kieserite, and herbicides, respectively. Based on the forecast, the amount
of fertilizer, kieserite, and herbicides used by 2030 is 2,395 ton, 656,601
The transport activity involves bringing the mature seeds from ton, and 2,064 L, respectively. Fig. 8 (a) shows the forecasting results for
nursery camps that located at several villages of the district to be planted fertilizer and kieserite, while Fig. 8 (b) is for herbicides. Meanwhile,
at the plantation areas. The oil palm trees are to replace the old un­ forecasting of the same substances used in the rearing of palm trees in
productive trees. Transportation mode used to send the seeds is 8-ton partner’s plantations is conducted using SLR method. Table 3 summa­
heavy duty vehicle (HDV) trucks that can haul 500 seeds in one de­ rizes MAPE and R-sq of the SLR models.
livery. The travel distance from nursery camp to the designated reju­
venated plantation sites and the number of seeds are summarized in 5. Estimation of FFB production in 2030
Table 2. The distance from the seedling sites to the rejuvenated locations
in 2030 is estimated to be 9,104 km. The last activity involved in a oil palm plantation is harvesting.
During harvesting, FFB are cut from the palm trees, collected, and
4. Estimation of oil palm rearing activity in 2030 hauled to the mill plant. Fig. 9 presents the predicted total production of
FFB from the company’s plantations, partners, loading ramp, and new
The palm trees rearing involves two major activities which are lands within 2022–2030. The forecasting adopts DES method based on
data of total FFB production from 2012 to 2021. FFB are harvested from
1,817 ha of company’s plantations, 7,960 ha of partner’s plantations,
Table 1 purchased from independent farmers at loading ramp, and including
Accuracy for the SLR models for palm oil nurtured in company’s plantation. those from 439 ha of new plantations that have been developed since
Palm oil related Parameter Estimated Predictor Accuracy
2021. The new plantations are expected to produce its first harvest in
activities (year) 2024. The DES method ends up with MAPE of 0.66%. The forecasting
MAPE R-sq
result of FFB production is estimated as 309,773 ton, increases from the
(%)
total production 298,446 ton in 2021. This increment is due to the in­
Nursery Fertilizer consumed Number of 0.24 0.907
crease in productive plantation areas in 2024.
seeds
Irrigation Pump Fuel Number of 0.33 0.802
used (diesel) seeds 6. Estimation of FFB transportation travel distance in 2030
Fungicide consumed Number of 0.11 0.985
seeds As the amount of FFB harvested from the partner’s plantation sites
Insecticide consumed Number of 0.51 0.881
seeds
change, the distance travelled to haul the FFB to the mill plant also

8
F. F et al. Journal of Cleaner Production 463 (2024) 142646

Fig. 7. (a). Forecasting results of fertilizer and diesel forecasting consumed in nursery. (b). Forecasting results of fungicides and insecticides consumed in nursery.

7. Estimation of the processing activities at palm oil mill in 2030

Table 2
Travel distance from nursery to the plantation areas.
FFB are transported by trucks to the palm oil mill. The processing
Nursery camp name Distance Number of matured Travel distance mill consists of stations for sterilization, threshing, pressing, seed sepa­
(village) (km) seeds (km)
ration and wastewater treatment (see Fig. 3). Fiber, shells, and empty
Nyogan 28 7,367 825 fruit bunches (EFB) are the solid waste that produced during palm oil
Pijoan 41 7,644 1,254
processing at the mill plant. Furthermore, a sizable volume of liquid
Berkah 33 6,110 807
Mekar Sari Nes 31 6,750 837 waste, known as palm oil mill effluent (POME) is concurrently produced.
Pompa Air 42 5,535 930 POME has the potential to be the source of environmental pollution
Panerokan 35 31,801 4,452 because it has organic compounds and solids that can disrupt the bal­
Total distance travelled 9,104 ance of the ecosystem. Components of POME include condensate water
from the boiling process, factory washing water, clarification waste­
changes according to the amount of FFB harvested in the areas. The FFB water from sludge separation, and the discharge from hydro-cyclone at
produced by the company’s plantations are carried from 4 different
areas while the ones from the partner’s plantation are carried from 28
Table 3
different areas. When calculating the distance, the types of vehicles are
Accuracy of the SLR models for the oil palm trees nurtured in partner’s plan­
assumed to be the same as those used in 2021, i.e. heavy-duty vehicle tation areas.
(HDV) of 8-ton capacity, light commercial vehicle (LCV) of 3.5-ton ca­
Palm oil related Parameter Predictor Accuracy
pacity, and HDV of 16-onne capacity. FFB produced from the company’s
activities estimated (year)
plantation are shipped by HDV trucks, while the ones from partner’s MAPE R-sq
plantation are carried by HDV and LCV trucks. (%)

The total travel distance for 8-ton HDV, 3.5-ton LVC, and 16-ton HDV Rearing in partner Fertilizer Plantation 0.53 0.788
is estimated at 2,059,938 km, 1,424,618 km, and 71,196 km, respectively. plantation consumed area
Kieserite Plantation 2.0 0.757
Compared to the total travel distance in 2021, the total distance in 2030
consumed area
increases by 6.5% and 10.45% for 8-ton HDV truck and 3.5 ton LCV, Herbicides Plantation 0.56 0.849
respectively. Meanwhile, the total travel distance for 16-ton HDV is very consumed area
much the same since the quantity of FFB transported is not that different.

Fig. 8. (a). Forecasting results of NPK fertilizer and kieserite consumed. (b). Forecasting results of herbicides consumed.

9
F. F et al. Journal of Cleaner Production 463 (2024) 142646

according to the upstream and downstream activities by 2030. This

estimated inventory will be used to calculate carbon footprint of the
palm oil industry in 2030.

4.2.3. Phase 3: life cycle impact assessment

According to the IPCC (Buendia et al., 2019) and the Ministry of
Energy and Mineral Resources greenhouse gas inventory report (ESDM,
2019), the emission related to specific activities are calculated using
Equations 1–11 with inputs of the estimated inventory as summarized in
Table 5 and the emission factors listed in Table 6. Direct emission is the
emission that is directly generated from a specific emission source, while
indirect emission is the emission generated as the result of activities due
to direct emission source. For instance, when dolomite is applied to palm
oil plants, direct emission occurs (the GHG emission is calculated using
factor listed on Table 6 grey colour). However, once the dolomite gets
into the soil, it reacts to form dolomite lime that also produces emission
(the GHG emission is calculated using factor listed on Table 6 cream
Fig. 9. the forecasting result of FFB production amount. colour). The emission produces by dolomite lime is indirect emission.
Table 7 summarizes the carbon footprint estimated by 2030. In this
the core station. About 60% of POME components come from the clar­ study, carbon footprint was estimated using OpenLCA software.
ification station, 35% come from the boiling station, while the remain­ In the life cycle impact assessment, carbon footprint is stated as the
ing 5% come from processing palm kernels into kernels (Ahmad et al., Global Warming Potential (GWP) category. The one used is GWP 100,
2020). Table 4 shows the estimate of EFB, POME, and fuels used in the which is based on the energy absorbed by the gases within 100 year
processing stations. Table 4 and Fig. 10 show that SLR forecasting are period (Lynch et al., 2020). Based on the calculation results, the total
excellent since their MAPE values are small and the R-sq values are very estimated carbon footprint in 2030 is 62,758,433.56 kg CO2eq. The
much close to 1. carbon footprint is accumulated from 18 emission sources. Table 7
shows that out of the 18 sources, insecticide is the one with the lowest
8. Estimation of finish products’ shipment activity in 2030 carbon footprint. Meanwhile, wastewater treatment (POME) is the
source with the highest carbon footprint with 3.037 × 107 kg CO2eq.
The main products of FFB processing are the CPO and the palm This emission source contributes to 48.44% (almost half) of the total
kernel oil (in short known as kernel). To predict the CPO and kernel carbon footprint generated by the palm oil company. Table 7 also shows
production, SLR method is used. Given the CPO and kernel production the top seven carbon footprint sources that emit CO2eq (1)% or more of
data from 2012 to 2021, the results show that the linear regression R-sq the total emissions.
coefficients for CPO and Kernel of FFB production are 0.956 and 0.794, The second contributor of carbon footprint after POME is NPK 15-15-
respectively. Fig. 11 displays the graphs of predicted CPO and kernel 15. NPK 15-15-15 is a type of fertilizers used during oil palm nursery and
production from palm oil mill. The total estimated CPO production in oil palm rearing before the plant produces its FFB. NPK 15-15-15 con­
2030 is 58,229 ton while the total kernel production is 15,076 ton. tributes more than 20% of the emissions. The next emissions’ source is 8-
The CPO and kernel produced will be shipped to fulfil their demand. ton HDV transportation truck. This HDV truck consumes diesel that
The products are transported from the production mill to a harbour. The leads to notorious emission. The 8-ton HDV Truck counts for almost 20%
trucks used to haul the CPO is the 16-ton HDV truck whereas for the of the total emissions. POME, NPK 15-15-15, and 8-ton HDV Truck are
kernel the 8-ton HDV truck is used. Since CPO production in 2030 is the biggest three emissions’ sources that contribute close to 90% of the
estimated to be 58,196 ton and the distance between the mill plant and total emissions. Meanwhile, the other four sources (NPK 16-16-16, 3.5-
Talang Duku harbour is 44 km, the total travel distance for 16-ton HDV ton LCV truck, heavy machine, and 16-ton HDV truck) contribute 1–5%
truck is estimated to be 256,208 km. Meanwhile the travel distance for of carbon footprint.
8-ton HDV truck that hauls the kernel is 66,335 km. compared to the The carbon footprint causes climate change that leave significant
total distance travelled in 2021, the total travel distance in 2030 for 16- impact to both ecosystems and human health (Priya et al., 2023; Tan
ton HDV and 8-ton HDV increases by 2.7% and 2.4%, respectively. et al., 2022; Rivas-Garcia et al., 2015). Besides being harmful to human
Table 5 summarizes the estimated inventory of inputs and output health, climate change and global warming also lead to extreme weather
that can cause a prolonged dry season or non-stop raining which results
in flooding. Floods renders the environment dirty. It even serves as a
Table 4 place for insects such as mosquitoes and flies to reproduce. The impacts
Forecasting accuracy of SLR models for CPO and Kernel processing. of greenhouse gasses, global warming, and extreme changes to human
Palm oil related Parameter estimated Predictor Accuracy health can be measured in terms of an index namely disability-adjusted
activities (year) life years (DALY). DALY is defined as the extra health damage due to the
MAPE R-sq
(%)
addition of 1 kg carbon dioxide emission. This index that developed by
Murray (1996) has been adopted by WHO. According to Tang et al.
CPO and kernel EFB Incineration The amount 0.68 0.946
(2019), DALY for country like Indonesia can be estimated by multiplying
processing of FFB
POME The amount 0.83 0.683 the carbon footprint with the value 1.3 * 10− 6. For the total carbon
of FFB footprint of 62,758,433.56 kg CO2eq, its DALY index is 81.59. DALY
Fibres as the The amount 0.69 0.927 index due to the palm oil carbon footprint according to each source
electricity fuel of FFB identified in this study are summarized in Table 8.
Shell as the electricity The amount 1.00 0.855
fuel of FFB
Engine room fuel The amount 1.24 0.747 4.2.4. Phase 4: life cycle interpretation
consumption of FFB From the total of 62,758,433.56 kg CO2eq, right column of Table 7
Heavy machine fuel The amount 0.78 0.865 informs about cumulative percentage of the GHG emissions. Values in
consumption of FFB
the column can be interpreted using Pareto chart, i.e. 80|20 Rule. The

10
F. F et al. Journal of Cleaner Production 463 (2024) 142646

Fig. 10. The forecasting results of: (a) the amount of fibers and shells. (b). the amount of EFB and POME. (c). diesel engine room and diesel heavy ma­
chine consumed.

Table 5
Summary of estimated O-LCA inventory in 2030.
No Activity Emission Sources (unit) Amount

1 Rejuvenation Excavator fuel (diesel) (liter) 4,605

Dolomite (kg) 119,866
2 Nursery Fertilizer (kg) 37,108
Irrigation pump fuel (diesel) 39,958
(liter)
Fungicides (liter) 71,283
Insecticide (liter) 106,740
3 Seeding shipment 8-ton HDV (diesel) (km) 9,104
4 Rearing at the company NPK Fertilizer (ton) 2,395.4
plantation Kieserite (ton) 656.6
Herbicide (liter) 2,064
5 Rearing at partner farmer NPK Fertilizer (ton) 9,149.41
plantation Kieserite (ton) 2,560.44
Herbicide (liter) 19,129
6 Company FFB shipment 8-ton HDV (diesel) (km) 436,250
Fig. 11. The quantity of palm oil final products forecasting results in 2030. 7 Partner Farmer FFB shipment 8-ton HDV (diesel) (km) 2,059,938
3.5-ton LCV (km) 1,424,618
8 Loading Ramp Transportation 16-ton HDV (km) 71,196
rule says roughly 80% of the effects come from 20% of causes. In this 8-ton HDV (km) 35,598
matter, 80|20 means 80% of carbon footprint are due to 20% of the 9 CPO & Kernel Processing EFB incineration (ton) 66,426
POME (m3) 160,191
sources. Fig. 7 shows that the 80% of carbon footprint contributors are
Fibres as electricity fuel (ton) 36,043
POME, NPK 15-15-15, and 8-ton HDV transportation truck. These three Shell as electricity fuel (ton) 452.96
emission sources, which contribute even more than 80% (i.e. 88.6%) of Engine room fuel (liter) 32,906
the total carbon footprint, are the ones should be addressed to ensure the Heavy machine fuel (liter) 57,610
palm oil industry is environmentally friendly. This finding is not sur­ 10 CPO shipment 16-ton HDV (km) 256,208
11 Kernel shipment 8-ton HDV (km) 66,335
prising. A study by Jamaludin et al. (2019) on carbon footprint showed

11
F. F et al. Journal of Cleaner Production 463 (2024) 142646

Table 6
The GHG emissions’ factors.

Table 7 Table 8
Estimated carbon footprint from 18 emission sources in 2030. Carbon footprint impact on human health in 2030.
Global Warming Potential 100 Disability-Adjusted Life Years (DALY)

No Emission Sources Carbon Footprint (kg % Cumulative No Emission Sources DALY No Emission Sources DALY
CO2eq) (%)
1 POME 39.52 10 Herbicide 0.091
1 POME 3.04E+07 48.44 48.44 2 NPK 15-15-15 17.16 11 Dolomite 0.074
2 NPK 15-15-15 1.32E+07 21.03 69.47 3 8-ton HDV 15.60 12 Excavator 0.054
3 8-ton HDV 1.20E+07 19.12 88.59 4 NPK 16-16-16 4.33 13 NPK 15-15-6 0.026
4 NPK 16-16-16 3.33E+06 5.31 93.90 5 3.5-ton LCV 1.53 14 Kieserite 0.006
5 3.5-ton LCV 1.18E+06 1.88 95.78 6 Heavy Machine 1.29 15 Incineration 0.002
6 Heavy Machine 9.91E+05 1.58 97.36 7 16-ton HDV 1.23 16 Biomass Electricity 0.001
7 16-ton HDV 9.45E+05 1.51 98.87 8 Irrigation Pump 0.55 17 Fungicides 0.0009
8 Irrigation Pump 4.25E+05 0.68 99.55 9 Engine Room 0.12 18 Insecticide 2.03E-06
9 Engine Room 9.07E+04 0.15 99.69 Total 81.59
10 Herbicide 7.01E+04 0.11 99.80
11 Dolomite 5.68E+04 0.09 99.89
12 Excavator 4.15E+04 0.07 99.96 that POME and diesel-based transportation are the two top sources of
13 NPK 15-15-6 2.03E+04 0.03 99.99 GHG emissions.
14 Kieserite 4.89E+03 0.01 99.99
15 Incineration 1.54E+03 <0.01 99.99
The finding of POME as the top source of carbon footprint in the palm
16 Biomass 9.09E+02 <0.01 ~100 oil industry in this study, i.e. 1 ton CPO produce 2.753 m3 of POME, is
Electricity consistent with the findings by Zainal et al. (2020). This resembles that
17 Fungicides 6.93E+02 <0.01 ~100 for every ton of CPO produced, 2.75 m3 of POME shall be generated. The
18 Insecticide 1.56E+00 100
finding is very much closed by Hazmi et al. (2016) and Saidu et al.
<0.01
Total Carbon Footprint 62,758,433.56
(2013) in which each ton of CPO contributes to 2.5 m3 of POME.
This study shows that fertilizer NPK 15-15-15 is the second source of
GHG emissions. This finding is parallel with Kusin et al. (2015) that
shows one kg of nitrogen fertilizer release 55 kg CO2eq. Another finding

12
F. F et al. Journal of Cleaner Production 463 (2024) 142646

shows that one km travel of 8-ton HDV truck contributes to 7.1 kg Data availability
CO2eq. Zahraee et al. (2022) showed that mode of transportation would
affect the amount of emissions. Using dynamic simulation model, they Data will be made available on request.
showed that utilizing larger capacity and size of transportation would
reduce the impact. References
This study has estimated that the amount of CPO the company will
produce in 2030 is 58,196 ton. Unfortunately, this amount will generate Ahmad, A.M., Zainal, B.S., Jamadon, N.H., Choong, T., Yaw, S., Abdullah, L.C., 2020.
Filtration analysis and fouling mechanisms of PVDF membrane for POME treatment.
62,758.433 ton CO2eq. On average a ton of CPO production will induce Journal of Water Reuse and Desalination 10 (3), 1–13. https://doi.org/10.2166/
1.08 ton CO2eq. This finding is lower than the estimation made by Rosa wrd.2020.101.
et al. (2022). Estimating GHG emissions at two palm oil plantations at Appavu, P., Madhavan, V.R., Jayaraman, J., 2019. Palm oil-based biodiesel as a novel
alternative feedstock for existing unmodified DI diesel engine. Int. J. Ambient
Central Kalimantan using LCA, Rosa et al. ended up with estimation of Energy 43, 222–228. https://doi.org/10.1080/01430750.2019.1636884.
2.72 and 2.25 kg CO2eq in producing 1 kg of palm oil. However, such Armah, F.A., Yawson, D.O., Pappoe, A.A.N.M., 2010. A system dynamics approach to
difference makes sense since Rosa et al. studied palm oil trees that explore traffic congestion and air pollution link in the City of Accra, Ghana.
Sustainability 2 (1), 252–265. https://doi.org/10.3390/su2010252.
cultivated on peat soil that generates its own carbon footprint. This Bappenas-LCDI, 2024. Pertanian. Badan Perencanaan Dan Pembangunan Nasional-Low
finding is also lower than Rosa et al. estimation since the study assumed Carbon Development Indonesia. https://lcdi-indonesia.id/grk-pertanian/. May
that age of the vehicles does not influence vehicles’ emission. Lam et al. 2024.
Bernardo, V., Fageda, X., Flores-Fillol, R., 2021. Pollution and congestion in urban areas:
(2019) estimated that avoiding palm oil tree cultivation on peat soil
the effects of low emission zones. Economics of Transportation 26–27, 100221.
areas will reduce the emission by 42%. Lam et al. estimated that CPO https://doi.org/10.1016/j.ecotra.2021.100221.
carbon footprint varies spatially from 0.7 to 26.0 ton CO2eq per ton of Bin, O., 2003. A logit analysis of vehicle emissions using inspection and maintenance
CPO. The finding of this study falls within the range. Assessing the testing data. Transport. Res. Transport Environ. 8 (3), 215–227. https://doi.org/
10.1016/S1361-9209(03)00004-X.
impact of palm oil certification on reducing GHG emissions, Schmidt Buendia, C.E., Tanabe, K., Kranjc, A., Baasansuren, J., Fukuda, M., Ngarize, S., Osako, A.,
and Rosa (2020) found that Roundtable on Sustainable Palm Oil (RSPO) Pyrozhenko, Y., Shermanau, P., Federici, S. (Eds.), 2019. 2019 Refinement to the
certification significantly reduces GHG emissions from 5.34 (3.34–8.16) 2006 IPCC Guidelines for National Greenhouse Gas Inventories. IPCC, Geneva,
Switzerland. https://www.ipcc-nggip.iges.or.jp/public/2019rf/index.html.
ton CO2eq per ton of CPO to 3.41 (2.61–4.48) CO2eq per ton of CPO. (Accessed 20 February 2022).
Byrne, R.F., 2012. Beyond traditional time-series: Using demand sensing to improve
5. Conclusion forecasts in volatile times. J. Bus. Forecas. 31 (2), 13–19.
Caserini, S., Pastorella, C., Gaifami, P., Ntziachristos, L., 2013. Impact of the dropping
activity with vehicle age on air pollutant emissions. Atmos. Pollut. Res. 4, 282–289.
This study offers a novel method to estimate carbon footprint for an https://doi.org/10.5094/APR.2013.031.
organization in the future, i.e. 2030. The method is a combination of Casolani, N., D’Eusanio, M., Liberatore, L., Raggi, A., Petti, L., 2022. Life cycle
assessment in the wine sector: a review on inventory phase. J. Clean. Prod. 379,
Organizational Life Cycle Assessment (O-LCA with double exponential 134404 https://doi.org/10.1016/j.jclepro.2022.134404.
smoothing (DES) and simple linear regression (SLR) forecasting Cucchi, M., Volpi, L., Ferrari, A.M., Garcia-Muiña, F.E., Settembre-Blundo, D., 2022.
methods. The forecasting methods are used to estimate the emission Industry 4.0 real-world testing of dynamic organizational life cycle assessment (O-
LCA) of a ceramic tile manufactureer. Environ. Sci. Pollut. Control Ser. https://doi.
sources’ inventory of all activities, including direct and indirect activ­
org/10.1007/s11356-022-20601-7.
ities. Even though DES and SLR forecasting methods are not the most Danilecki, K., Eliasz, J., Smurawski, P., Stanek, W., Szlęk, A., 2021. Modeling inventory
recent developed methods, but they give excellent estimation results and environmental impacts of car maintenance and repair: a case study of Ford
since the MAPE results are less than 10%. In this study, the forecasting Focus passenger car. J. Clean. Prod. 315, 128085 https://doi.org/10.1016/j.
jclepro.2021.128085.
was conducted using Microsoft Excel spreadsheets, while carbon foot­ Deeney, M., Green, R., Yan, X., Dooley, C., Yates, J., Rolker, H.B., Kadiyala, S., 2023.
print estimation is using OpenLCA software. The carbon footprint result J. Clean. Prod. 397, 136567 https://doi.org/10.1016/j.jclepro.2023.136567.
is 62,758,433.56 kg CO2eq. On average a ton of CPO produced will Dohong, A., Aziz, A.A., Dargusch, P., 2018. Carbon emissions from oil palm development
on deep peat soil in central kalimantan Indonesia. Anthropocene 22, 31–39. https://
generate 1.08 ton CO2eq. Three main emission sources are identified, doi.org/10.1016/j.ancene.2018.04.004.
namely palm oil liquid waste (POME), fertilization, and transportation. El Joumri, L., Labjar, N., Dalimi, M., Harti, S., Dhiba, D., El Messaoudi, N., Bonnefille, S.,
Today, palm oil is an essential inevitable human consumption. El Hajjaji, S., 2023. Life cycle assessment (LCA) in the olive oil value chain:
descriptive review. Environmental Development 45, 100400. https://doi.org/
Nevertheless, it releases greenhouse gases (GHGs) that endangers living 10.1016/j.envdev.2022.100800.
being on earth. Therefore, future study should address the ways to Ernawati, H., Saputra, A., Alamsyah, Z., Napitupulu, D.M., Yanita, M., Fauzia, G., 2021.
reduce carbon footprint to ensure the sustainability of this industry. Analysis of independent oil palm farming income after replanting in muaro Jambi
District. IOP Conf. Series: Earth and Environmental Science 1–8. https://doi.org/
Many oil palm plantations owned by private entities that pursuing profit 10.1088/1755-1315/782/3/032055.
as their main business objective. Reducing carbon footprint on their ESDM, 2019. Inventarisasi Emisi GRK Sektor Energi. https://www.esdm.go.id/assets
business activities means adding more cost. Future research may also /media/content/content-inventarisasi-emisi-gas-rumah-kaca-sektor-energi-tahun-
2020. (Accessed 20 February 2022).
deal with multi-criteria strategy to encourage their involvement. The
Farizal, F., Dachyar, M., Taurina, Z., Qaradhawi, Y., 2021. Disclosing fast moving
criteria may include economic, technical, and social as addition to consumer goods demand forecasting predictor using multi linear regression.
environmental. Another possible direction is on the forecasting method Engineering and Applied Science Research 48 (5), 627–636. https://doi.org/
itself. Future study may incorporate more advanced methods such as 10.14456/easr.2021.64.
Farizal & Septia, W.E., 2015. Life assessment analysis of Photovoltaic system in
multilinear regression and artificial neural networks that allow to use Indonesia. In: Proceeding of the 14th International Conference on QIR (Quality in
multiple predictors. Research), pp. 1220–1224. ISSN: 1411-1284.
Farnum, N.R., Stanton, L.-V.W., 1989. Quantitaive Forecasting Method. PWS-KENT,
United States, pp. 22–25.
CRediT authorship contribution statement Forin, S., Mikosch, N., Berger, M., Finkbeiner, M., 2019. Organizational water footprint :
a methodological guidance. Int. J. Life Cycle Assess. 25, 403–422. https://doi.org/
Farizal F: Writing – review & editing, Writing – original draft, 10.1007/s11367-019-01670-2.
Fotios, P., Konstantinos, N., Georgios, P.S., Vassilis, A., 2013. Empirical heuristics for
Validation, Supervision, Methodology, Conceptualization. Trisha improving intermittent demand forecasting. Ind. Manag. Data Syst. 113, 683–696.
Amanda: Writing – original draft, Visualization, Investigation, Formal Francke, I.C.M., Castro, J.F.W., 2013. Carbon and water footprint analysis of a soap bar
analysis, Data curation. Muhammad Dachyar: Supervision. Zainura produced in Brazil by Natura Cosmetics. Water Resour. Ind. 1–2, 37–48. https://doi.
org/10.1016/j.wri.2013.03.003.
Zainon Noor: Writing – review & editing, Validation, Supervision. Gennitsaris, S., Sagani, A., Sofianopoulou, S., Dedoussis, V., 2023. Integragted LCA and
DEA approach for circular economy-driven performance evaluation of wind turbine
Declaration of Competing interest end-of-life treatment options. Appl. Energy 339, 120951. https://doi.org/10.1016/j.
apenergy.2023.120951.

We have no financial and conflicts of interest to declare.

13
F. F et al. Journal of Cleaner Production 463 (2024) 142646

George, K.M., Park, N., Yang, Z., 2015. A reliability measure for time series forecasting Martinez, S., Bessou, C., Hure, L., Guilbot, J., Hélias, A., 2017. The impact of palm oil
predictor. IFAC-PapersOnLine 48, 850–855. feedstock within the LCA of a bio-sourced cosmetic cream. J. Clean. Prod. 145,
Gilliland, M., 2010. The Business Forecasting Deal: Exposing Myths, Eliminating Bad 348–360. https://doi.org/10.1016/j.jclepro.2017.01.042.
Practices, Providing Practical Solutions. SAS Institute Inc. McAvoy, S., Grant, T., Smith, C., Bonti, N.C.K., 2021. Combining life cycle assessment
Greene, D.L., Liu, J., Khattak, A.J., Wali, B., Hopson, J.L., Goeltz, R., 2017. How does on- and system dynamics to improve impact assessment. J. Clean. Prod. 315, 128060
road fuel economy vary with vehicle cumulative mileage and daily use. Transport. https://doi.org/10.1016/j.jclepro.2021.128060.
Res. Transport Environ. 55, 142–161. https://doi.org/10.1016/j.trd.2017.06.004. !"count(.//sb:host[1]//child::*//sb:date)">Mitchell, A., Dodge, B.H., Kruzic, P.G.,
Hamada, H.M., Jokhio, G.A., al-Attar, A.A., Yahaya, F.M., Muthusamy, Kh, Humada, A. Miller, D.C., Schwartz, P., Suta, B.E., . Handbook of forecasting techniques. In:
M., Gul, Y., 2020. The use of palm oil clinker as a sustainable construction material: National Technical Information Services. US Department of Commerce, Springfield,
a review. Cement Concr. Compos. 106, 103447 https://doi.org/10.1016/j. VA, pp. 283–292.
cemconcomp.2019.103447. Montgomery, D.C., Jennings, C.L., Kulahci, M., 2008. Introduction to Time Series
Hashim, Z., Subramaniam, V., Harun, M.H., Kamarudin, N., 2018. Carbon footprint of oil Analysis and Forecasting. Wiley & Sons Publication, Hoboken, NJ., pp. 1–49
palm planted on peat in Malaysia. Int. J. Life Cycle Assess. 23 (6), 1201–1217. Murray, C.J.L., 1996. Rethinking DALYs. In: The Globar Burden of Disease. A
https://doi.org/10.1007/s11367-017-1367-y. Comprehensive Assessment of Mortality and Disability from Diseases, Injuries, and
Hauschild, M.Z., Rosenbaum, R.K., Olsen, S.I., 2018. Life Cycle Assessment. Springer. Risk Factors in 1990 and Projected to 2020. Harvard University Press, Cambridge.
Hazmi, A., Desmiarti, R., Waldi, E.P., Emeraldi, P., 2016. Preliminary study on treatment Notarnicola, B., Tassielli, G., Renzulli, P.A., Chirulli, P., 2023. Organizational life cycle
of palm oil mill effluent by sand filtration-dielectric barrier discharge system. assessment of a corporate group. Cleaner environmental Systems 9, 100125. https://
Journal of Engineering and Technological Sciences 48 (1), 21–30. https://doi.org/ doi.org/10.1016/j.cesys.2023.100125.
10.5614/j.eng.technol.sci.2016.48.1.3. Noya, L.I., Vasilaki, V., Stojceska, V., González-García, S., Kleynhans, C., Tassou, S.,
ISO 14040, 2006. Environmental Management-Life Cycle Assessment-Principles and Moreira, M.T., Katsou, E., 2018. An environmental evaluation of food supply chain
Framework. The International Organization for Standardization. using life cycle assessment: a case study on gluten free biscuit products. J. Clean.
Hesser, F., 2015. Environmental advantage by choice: Ex-ante LCA for a new Kraft pulp Prod. 170, 451–461. https://doi.org/10.1016/j.jclepro.2017.08.226.
fibre reinforced polypropylene composite in comparison to reference materials. Oda, H., Noguchi, H., Fuse, M., 2022. Review of life cycle assessment for automobiles: a
Composite Part B 79, 197–203. https://doi.org/10.1016/j. meta-analysis-based approach. Renew. Sustain. Energy Rev. 159, 112214 https://
compositesb.2015.04.038. doi.org/10.1016/j.rser.2022.112214.
Izzah, N., Aziz, H.A., Hana, M.M., 2020. Life cycle analysis of biogas production from Pereira, S.R., Coelho, M.C., 2015. Can nanomaterials be a solution for application on
anaerobic digestion of palm oil mill effluent. Renew. Energy 145, 847–857. https:// alternative vehicles?- A review paper on life cycle assessment and risk analysis. Int.
doi.org/10.1016/j.renene.2019.06.084. J. Hydrogen Energy 40, 4969–4979. https://doi.org/10.1016/j.
Jamaludin, N.F., Muis, Z.A., Hashim, H., 2019. An integrated carbon footprint ijhydene.2014.12.132.
accounting and sustainability index for palm oil mills. J. Clean. Prod. 225, 496–509. PP, 2021. Peraturan presiden nomor 98 tahun 2021 tentang penyelenggaraan nilai
https://doi.org/10.1016/j.jclepro.2019.03.312. ekonomi karbon untuk mencapai target kontribusi yang ditetapkan secara nasional
Jeswani, H.K., Hellweg, S., Azapagic, A., 2018. Accounting for land use, biodiversity and dan pengendalian emisi gas rumah kaca dalam pembangunan nasional. Presidensial
ecosystem services in life cycle assessment: impacts of breakfast cereals. Sci. Total Decree Number 98 Year of 2021 on Carbon Economic Value Management to Reach
Environ. 645, 51–59. https://doi.org/10.1016/j.scitotenv.2018.07.088. Nasional Target and Greenhouse Gas Reduction Emissions. https://peraturan.bpk.
Jungbluth, N., Keller, R., König, A., 2016. One TWO WE – life cycle managemant in go.id/Details/187122/perpres-no-98-tahun-2021. (Accessed 16 September 2023).
canteens together with suppliers, customers and guests. Int. J. Life Cycle Assess. 21, Pranolo, S.H., Waluyo, J., Putro, F.A., Adnan, M.A., Kibria, MdG., 2023. Gasification
646–653. https://doi.org/10.1007/s11367-015-0982-8. process of palm kernel shell to fuel gas: pilot-scale experiment and life cycle analysis.
Kementan, 2020. Statistical of national leading estate Crops commodity. Directorate Int. J. Hydrogen Energy 48, 2835–2848. https://doi.org/10.1016/j.
General of Estate Crops. https://drive.google.com/file/d/1ZpXeZogAQYfClNBOgV ijhydene.2022.10.066.
LhYi8X_vujJdHx/view. Priya, A.K., Muruganandam, M., Rajamanickam, S., Sivarethinamohan, S., Gaddam, M.K.
Kim, S., Kim, H., 2016. A new metric of absolute percentage error for intermittent R., Velusamy, P., Gomathi, R., Ravindiran, G., Gurugubelli, T.R., Muniasamy, S.K.,
demand forecasts. Int. J. Forecast. 32, 669–679. 2023. Impact of climate change and anthropogenic activities on aquatic ecosystem –
Khatiwada, D., Palmén, C., Silveira, S., 2021. Evaluating the palm oil demand in a review. Environ. Res. 238, 117233 https://doi.org/10.1016/j.
Indonesia: production trends, yields, and emerging issues. Biofuels 12 (2), 135–147. envres.2023.117233.
https://doi.org/10.1080/17597269.2018.1461520. Queiroz, A.G., Franca, L., Ponte, M.X., 2012. The life cycle assessment of biodiesel from
Kjaer, L.L., Pigosso, D.C.A., Mcaloone, T.C., Birkved, M., Louise, A., 2018. Guidelines for palm oil (“dendê”) in the Amazon. Biomass Bioenergy 36, 50–59. https://doi.org/
evaluating the environmental performance of product/service-systems through life 10.1016/j.biombioe.2011.10.007.
cycle assessment. J. Clean. Prod. 190, 666–678. https://doi.org/10.1016/j. Resta, B., Gaiardelli, P., Pinto, R., Dotti, S., 2016. Enhanching environmntal management
jclepro.2018.04.108. in the textile sector: an organisational-life cycle assessment approach. J. Clean. Prod.
Klöpffer, W., Curran, M.A., 2017. Goal and Scope Definition in Life Cycle Assessment. 135, 620–632. https://doi.org/10.1016/j.jclepro.2016.06.135.
Springer. Rimano, M., Simboli, A., Taddeo, R., Del Grosso, M., Raggi, A., 2021. The environmental
Kumar, S., Pal, A., 2023. Life cycle analysis of biodiesel derived from fresh water impact of organizations: a pilot test from the packaging industry based on
microalgae and Karanja. Sustain. Energy Technol. Assessments 56, 103082. https:// organizational life cycle assessment. Sustainability 13 (20), 1–18. https://doi.org/
doi.org/10.1016/j.seta.2023.103082. 10.3390/su132011402.
Kusin, F.M., Akhir, N.I.M., Yusuff, F.M., Awang, M., 2015. The impact of nitrogen Rimano, M., Simboli, A., Taddeo, R., Raggi, A., 2019. Life cycle approaches for the
fertilizer use on greenhouse gas emissions in an oil palm plantation associated with environmental impact assessment of organizations: defining the state of the art.
land use change. Atmósfera 28 (4), 243–250. https://doi.org/10.20937/ Adm. Sci. 9 (4), 94. https://doi.org/10.3390/admsci9040094.
ATM.2015.28.04.03. Ritchie, H., Roser, M., 2021. Forests and deforestation. OurWorldInData.org. https://o
Lam, W.Y., Kulak, M., Sim, S., King, H., Huijbregts, M.A.J., Chapin-Kramer, R., 2019. urworldindata.org/forests-and-deforestation. (Accessed 30 March 2023).
Greenhouse gas footprints of palm oil production in Indonesia over space and time. Rivas-Garcia, P., Botello-Álvarez, J.E., Seabra, J.E.A., Walter, A.C.S., Estrada-
Sci. Total Environ. 688, 827–837. https://doi.org/10.1016/j.scitotenv.2019.06.377. Baltazar, A., 2015. Environmental implications of anaerobic digestion for manure
Lee, U., Cai, H., Ou, L., Benavides, P.T., Ang, Y., Wang, M., 2023. Life cycle analysis of management in dairy farms in Mexico: a life cycle perspective. Environ. Technol. 36
gasification and Fischer-Tropsch conversion of munipal solid waste for (17), 2198–2209. https://doi.org/10.1080/09593330.2015.1024758.
transportation fuel production. J. Clean. Prod. 382, 135114 https://doi.org/ Rivera-Méndez, Y.D., Rodriguez, D.T., Romero, H.M., 2017. Carbon footprint of the
10.1016/j.jclepro.2022.135114. production of oil palm (Elaeis guineensis) fresh fruit bunches in Colombia. J. Clean.
Luisetti, T., Ferrini, S., Grilli, G., Jickells, T.D., Kennedy, H., Kröger, S., Lorenzoni, I., Prod. 149, 743–750. https://doi.org/10.1016/j.jclepro.2017.02.149.
Milligan, B., van der Molen, J., Parker, R., Pryce, T., Turner, R.K., Tyllianakis, E., Rosa, M.D., Schmidt, J., Pasang, H., 2022. Industry-driven mitigation measures can
2020. Climate action requires new accounting guidance and governance frameworks reduce GHG emissions of palm oil. J. Clean. Prod. 365, 132565 https://doi.org/
to manage carbon in shelf seas. Nat. Commun. 11 (1) https://doi.org/10.1038/ 10.1016/j.jclepro.2022.132565.
s41467-020-18242-w. Nature Research. Rum, I.A., Tukker, A., de Koning, A., Yusuf, A.A., 2022. Impact assessment of the EU
Lynch, J., Cain, M., Pierrehumbert, R., Allen, M., 2020. Demonstrating GWP *: a means import ban on Indonesian palm oil: using environmental extended multi-scale MRIO.
of reporting warming-equivalent emissions that captures the contrasting impacts of Sci. Total Environ. 853, 158695 https://doi.org/10.1016/j.scitotenv.2022.158695.
short- and long-lived climate pollutants. Environ. Res. Lett. 15, 044023. Sabli, N.S.M., Noor, Z.Z., Kanniah, K.A.P., Kamaruddin, S.N., Rusli, N.M., 2017.
Manzardo, A., Loss, A., Mazzi, A., Scipioni, A., 2015. Organizational life cycle assessment Developing a methodology for water footprint of palm oil based on a methodological
(O-LCA): methodological issues and case studies in the beverage-packaging sector. review. J. Clean. Prod. 146, 173–180. https://doi.org/10.1016/j.
In: Muthu, S.S. (Ed.), Environmental Footprints of Packaging. Springer, Singapore, jclepro.2016.06.149.
pp. 47–63. Saidu, M., Yuzir, A., Salim, M.R., Salmiati, Azman, S., Abdullah, N., 2013. Influence of
Martinez-Blanco, J., Forin, S., Finkbeiner, M., 2018. Launch of a new report: “Road palm oil mill effluent as inoculum on anaerobic digestion of cattle manure for biogas
testing organizational life cycle assessment around the world: applications, production. Bioresour. Technol. 141, 174–176.
experiences and lessons learned.”. Int. J. Life Cycle Assess. 23, 159–163. https://doi. Sawangkeaw, R., Teeravitud, S., Piumsomboon, P., Ngamprasertsith, S., 2012. Biofuel
org/10.1007/s11367-017-1409-5. production from crude palm oil with supercritical alcohols: comparative LCA studies.
Martínez-Blanco, J., Inaba, A., Finkbeiner, M., 2015b. Scoping organizational Bioresour. Technol. 120, 6–12. https://doi.org/10.1016/j.biortech.2012.06.014.
LCA—challenges and solutions. Int. J. Life Cycle Assess. 20 (6), 829–841. https:// Schmidt, J., Rosa, M.D., 2020. Certified palm oil reduces greenhouse gas emissions
doi.org/10.1007/s11367-015-0883-x. compared to non-certified. J. Clean. Prod. 277, 124045 https://doi.org/10.1016/j.
Martinez-Blanco, J., Finkbeiner, M., Inaba, A., 2015a. Guidance on Organizational Life jclepro.2020.124045.
Cycle Assessment. UNEP/SETAC Life Cycle Initiative.

14
F. F et al. Journal of Cleaner Production 463 (2024) 142646

Shahbandeh, M., 2023. Palm oil usage worldwide 2015/16-2022/23. https://www. Vazifeh, Z., Bensebaa, F., Shadbahr, J., Gonzales-Calienes, G., Mafakheri, F., Benali, M.,
statista.com/statistics/274127/world-palm-oil-usage-istribution/#statistic Ebadian, M., Vézina, P., 2023. Forestry based products as climate change solution:
Container. (Accessed 29 March 2023). integrating life cycle assessment with techno-economic analysis. J. Environ. Manag.
Sou, M., Domenech, T., 2022. Integrating LCA and blockchain technology to promote 330, 117197 https://doi.org/10.1016/j.jenvman.2022.117197.
circular fashion - a case study of leather handbags. J. Clean. Prod. 373, 133557 Vieira, A.W., Rosso, L.S., Demarch, A., Pasini, Dj, Ruzza, S.P., Arcano, S., Ribeiro, M.J.,
https://doi.org/10.1016/j.jclepro.2022.133557. Angioletto, E., 2023. Life cycle assessment in the ceramic tyle industry: a review.
Sparrevik, M., Utstøl, S., 2019. Assessing life cycle greenhouse gas emissions in the Journal of Research and Technology 23, 3904–3915. https://doi.org/10.1016/j.
Norwegian defence sector for climate change mitigation. J. Clean. Prod. 248, 119196 jmrt.2023.02.023.
https://doi.org/10.1016/j.jclepro.2019.119196. Vining, G.G., 1998. Statistical Methods for Engineers. Duxbury Press. Brooks/Cole
Su, G., Zulkifli, N.W.M., Ong, H.C., Ibrahim, S., Cheah, M.Y., Zhu, R., Bu, Q., 2023. Co- Publishing Company, Pacific Grove, CA.
pyrolysis of medical protective clothing and oil palm wastes for biofuel: Wang, P., Zhang, R., Sun, S., Gao, M., Zheng, B., Zhang, D., Zhang, Y., Carmichael, G.R.,
experimental, techno-economic, and environmental analyses. Energy 273, 127221. Zhang, H., 2023. Aggravated air pollution and health burden due to traffic
https://doi.org/10.1016/j.energy.2023.127221. congestion in urban China. Atmos. Chem. Phys. 23, 2983–2996. https://doi.org/
Syahminar, Harahap, E.M., Rauf, A., Jamil, A., 2020. Effect of ameliorant material 10.5194/acp-23-2983-2023.
application on leaf nutrient uptake of oil palm (Elaeis Guineensis Jacq) seedings in Wang, Y., Xing, Z., Zhang, H., Wang, Y., Du, K., 2022. On-road mileage-based emission
peat planting media in the main nursery. International Journal of Multi Science 1 factors of gaseous pollutants from bi-fuel taxi fleets in China: the influence of fuel
(5), 70–79. type, vehicle speed, and accumulated mileage. Sci. Total Environ. 819, 151999
Tam, V.W.Y., Zou, Y., Shen, L., Le, K.N., 2023. Optimal BIM and LCA integration https://doi.org/10.1016/j.scitotenv.2021.151999.
approach for embodied environmental impact assessment. J. Clean. Prod. 385, Wei, F., Tan, Q., Dong, K., Li, J., 2022. Revealing the feasibility and environmental
135605 https://doi.org/10.1016/j.jclepro.2022.135605. benefits of replacing disposable plastic tableware in aviation catering: an AHP-LCA
Tan, X., Liu, Y., Dong, H., Xiao, Y., Zhao, Z., 2022. The health consequences of integrated study. Resour. Conserv. Recycl. 187, 106615 https://doi.org/10.1016/j.
greenhouse gas emissions: a potential pathway. Environ. Geochem. Health 44, resconrec.2022.106615.
2955–2974. https://doi.org/10.1007/s10653-021-01142-3. Wursthorn, S., Feifel, S., Walk, W., Patyk, A., 2010. An environmental comparison of
Tang, L., Furushima, Y., Honda, Y., Hasegawa, T., Itsubo, N., 2019. Estimating human repair versus replacement in vehicle maintenance. Transport. Res. Transport
health damage factors related to CO2 emissions by considering updated climate- Environ. 15 (6), 356–361. https://doi.org/10.1016/j.trd.2010.02.011.
related relative risks. Int. J. Life Cycle Assess. 24, 1118–1128. https://doi.org/10. Yilmaz, B., Ağagündüz, D., 2022. Fractionated palm oils: emerging roles in the food
1007/s11367-018-1561-6. industry and possible cardiovascular effects. Crit. Rev. Food Sci. Nutr. 62 (7),
Toniolo, S., Marson, A., Fedele, A., 2023. Combining organizational and product life 1990–1998. https://doi.org/10.1080/10408398.2020.1869694.
cycle perspective to explore the environmental benefits of steel slag recovery Zainal, B.S., Ahmad, M.A., Danaee, M., Jamadon, N., Mohammad, N.S., Ibrahim, S.,
practices. Sci. Total Environ. 867, 161440 https://doi.org/10.1016/j. 2020. Integrated system technology of POME treatment for biohydrogen and
scitotenv.2023.161440. biomethane production in Malaysia. Appl. Sci. 10 (3), 951. https://doi.org/10.3390/
Uning, R., Latif, M.T., Othman, M., Juneng, L., Hanif, N.M., Nadzir, M.S.M., Maulud, K.N. app10030951.
A., Jaafar, W.S.W.M., Said, N.F.S., Ahamad, F., Takriff, M.S., 2020. A review of Zahraee, S.M., Golroudbary, S.R., Shiwakoti, N., Stasinopoulos, P., 2022. Palm oil
southeast Asian oil palm and its CO2 fluxes. Sustainability 12 (12), 1–15. https:// biomass global supply chain: environmental emissions vs technology development of
doi.org/10.3390/su12125077. maritime transportation. Procedia CIRP 105, 817–822.
US EPA, 2006. Life Cycle Assessment: Principles and Practice. Scientific Applications
International Corporation (SAIC), Reston, VA, USA. Doc. #: EPA/600/R-06/060,
May 2006.

15