E-ISSN: 2528-388X
P-ISSN: 0213-762X
INERSIA
Vol.
No.
May 2025 Mode Choice and Spatial Distribution of Coal Transport in Jambi.
Indonesia Nurman Nugroho*.
Sigit Priyanto.
Bambang Triatmodjo Department of Civil and Environmental Engineering.
Universitas Gadjah Mada.
Yogyakarta 55281.
Indonesia
ABSTRACT
Keywords:
Generalized cost Mode choice Multinomial logit Spatial Coal is an important commodity for Jambi Province.
Based on data from the Central Bureau of Statistics, in the first quarter of 2022, the value of coal exports reached 10.
75% of the total export value.
Apart from being exported, coal commodity is also used for domestic purposes.
The transport used at this time is only through public roads.
The use of these routes results in many problems such as traffic accidents, congestion, and social conflicts.
Therefore, the number of vehicles allowed to pass on public roads is limited.
This has resulted in the production target plan not being achieved, only 17.
3 million tonnes out of 40 million tonnes in 2022.
Hence the need for other routes for transport such as rivers, special roads, and railways.
This research is intended to analyse coal transport trips using trucks, barges and railways.
This research discusses the closest route, mode selection, and transport costs based on the distance travelled and the travel time of each mode.
The search for the closest travel route is done by spatial analysis with Network Analyst on ArcGIS.
Mode choice was analysed using the multinomial logit method.
Meanwhile, transport costs are calculated based on the principle of generalised Modelling results on the selection of mode, during the rainy season the most efficient mode is barging with a selected probability of 44%, while rail 28% and transport by truck 27%.
During the dry season, the probability of transport by railway is 93%, barge 4% and truck 3%.
The results indicate that the most efficient mode during the rainy season is barging and during the dry season transport by railway.
This is an open access article under the CCAeBY license.
Introduction As the demand for coal in the global market continues to increase, coal mining capacity has increased rapidly.
This was marked by the establishment of a coal production target plan of 40 million tonnes in 2022 by the Ministry of Energy and Mineral Resources for Jambi Province.
However, according to the Ministry of Energy and Mineral Resources, the target plan was not achieved.
Coal production in Jambi Province was only able to reach 17.
million tonnes in 2022.
The failure to achieve the target is caused by problems that occur in the transport or distribution process.
During this time, the transport of coal from mining areas only relies on public roads.
Coal distribution through public roads often causes problems such as accidents involving coal transport vehicles and other road user vehicles.
Severe congestion, pavement damage and social conflict are also other consequences of transporting coal via public roads.
This means that coal transport in Jambi Province does not meet the criteria for effectiveness and efficiency as stated in the National Transport System (SISTRANAS).
Therefore, the Jambi *Corresponding author.
E-mail: nurman.
nugroho@mail.
https://doi.
org/10.
21831/inersia.
Received 23th July 2023.
Revised 30th May 2025.
Accepted 12th June 2025 Available online 13th June 2025 Provincial Government plans to build a special road for coal transport and build Tenam Port as a supporting infrastructure for the movement of coal commodities.
order to organise effective and efficient coal transport as required in the National Transport System (SISTRANAS), there is a need for a comprehensive analysis of the That is why this research was conducted by modelling coal transport distribution in Jambi Province.
Several studies with various models have been conducted to determine the optimal freight transport journey.
Modelling in the aspect of freight transport often uses multinomial logit as its mode choice model.
As has been done in previous study .
who used a multinomial logit model with travel time, travel costs, and climate as determining variables to determine the choosing of crude palm oil (CPO) transport modes.
Research with a similar model .
, which determined the modes choice of freight transport on the island of Java using travel costs, waiting time, and reliability as the determining variables.
Reference .
in their research, used the lowest cost concept and .
used the closest transport network model INERSIA.
Vol.
No.
May 2025 Nurman Nugroho, et.
to model coal transportation without any other factors as decision-making parameters by the owner of the goods.
This research was conducted with the intention of determining the optimal mode choice based on the optimal generalised cost and the closest travel route .
The model used is multinomial logit with travel cost, travel time and climate as determining attributes.
saving (VFTTS) can be defined as the monetary value derived from the reduction in units of time required to move goods between two locations .
In a previous study, estimating the value of savings with meta-models on freight transport in several countries.
The VFTTS value in Indonesia obtained from the study is summarised in Table Methods 3 Mode Choice In general, the modelling in this study includes several The research began with spatial modelling, generalised cost analysis, and mode choice modelling.
Each modelling and analysis use various software as tools.
There are various models that can be used to estimate the best mode choice.
The multinomial logit model was chosen to determine the mode of transport in this study.
The model has been widely used for modelling passenger and freight Table 3 summarises the data and attributes required in modelling mode choice.
1 Spatial Modelling The spatial model is intended to illustrate where the areas that act as generators of travel and destinations for coal transport travel.
As in the previous research .
, the spatialbased transportation modelling process is assisted by an ArcGIS software that has been equipped with a special tool for transportation network analysis called Network Analyst .
The data required for spatial-based transport modelling include transport networks, administrative boundaries, and other land use information.
Details of the data required for spatial-based transport modelling can be seen in Table 1.
2 Generalised Cost Analysis The second stage of modelling in this study was to determine and select the most optimal transport modes based on generalised costs, referring to the opinion expressed by .
The generalised cost analysis uses Microsoft Office Excel number crunching software.
The generalised cost calculation formula can be written as Equation 1.
yaycnycyco = ycNyaycnycyco ycNycNycnycyco ycOycuycN yu where yaycnycyco is the overall generalised cost characterised by currency units (R.
ycNyaycnycyco is the vehicle travel rate per unit distance (Rp/k.
Costs incurred based on the length of the journey are expressed by travel time ( ycNycNycnycyco ) using time units, as well as the time value of the goods transported ( ycOycuycN ) using currency units per unit time.
The other components of travel costs are denoted by .
Since time cannot be owned and traded, the value of travel time savings should be introduced .
The value of travel time based on savings is represented by a monetary value, whereby savings in travel time in a particular context can be compensated with monetary savings.
In the aspect of freight transport known as value of freight travel time Data Table 1.
Spatial data for modelling Sources Inventory of transport networks and Departemen Perhubungan.
Departemen Pekerjaan Umum dan Perumahan Rakyat Mining area Dinas Energi dan Sumber Daya Mineral Provinsi Jambi.
Regional administration map.
Hydrology map.
Land Badan Informasi Geospasial Local regulations of transport and land use Badan Perencanaan dan Pengembangan Daerah Provinsi Jambi.
Dinas Perhubungan Provinsi Jambi Table 2.
Value of freight travel time in Indonesia .
VoT (USD per tonne/hou.
Route Carriers Shippers Road Railway Air Sea Inland Waterway Table 3.
Mode choice modelling data and attributes Data/Attributes Data Source Speed Truck: Observation.
Railway:
Literature review .
Barge:
Literature review .
Distance Spatial modelling Travel time Spatial modelling and calculation Travel cost Truck: calculation.
Railway:
Literature review .
Barge:
Climate Balai Wilayah Sungai Sumatera IV Nurman Nugroho, et.
INERSIA.
Vol.
No.
May 2025 Multinomial logit model can be written as Equation 2.
ycEycu .
= Ocycu exp Vi yc=1.
xp V .
modelling also considers climatic conditions in addition to the general factors mentioned.
1 Travel Cost Pn is the probability that alternative i is chosen by individual n in the choice set Cn.
The multinomial logit model is used by most studies because it is a simple model.
With this consideration, the multinomial logit model was used in this study.
The travel costs of each mode are obtained based on vehicle operating cost calculations and literature studies as discussed in the previous chapter.
The travel costs of each mode are summarised in Table 4.
4 Modelling Procedures Modes Max Capacity Broadly speaking, the modelling in this study starts with the preparation of spatial data.
Then, a geographic information system (GIS) simulation was conducted using ArcGIS software with transport attributes .
ehicle speed, trip limit.
as input.
The simulation results .
are in the form of an origin-destination matrices with the closest travel route and its attributes .
istance and travel tim.
With the new travel attributes obtained, the generalised cost of each mode can be calculated to determine the most optimal transport mode.
An overview of the modelling procedure can be seen in Figure 1.
Truck Barge Railway 15 tons 000 - 12.
000 tons 50 tons/car Table 4.
Travel cost of coal transport Cost (Rp/tonne k.
2 Spatial Distribution of Coal Transport As seen in Figure 2, the trips originated from 100 coal mine sites and were destined to three ports.
During the rainy season, the entire port is assumed to be able to service coal Meanwhile, during the dry season, there are only two ports that can handle shipments.
Coal transport by trucks over public road Result and Discussion Travel modelling studies usually consider various influencing factors.
Factors that are commonly considered for freight transport trips are distance, travel time, freight fares, as well as several other factors.
In the aspect of freight transport, sometimes the influence factors are also inherent along with the characteristics of the goods or the characteristics of the transport mode itself.
Such as previous research .
, this study on coal transport Figure 3Figure illustrates, the most common shortest travel route from the mine is to Tenam Port.
A total of 77 out of 100 shortest travel routes are to Tenam Port.
The shortest travel route to the special terminal in Tebing Tinggi is 15 out of 100 routes.
The other part is a route with the destination of Talang Duku Port, namely 8 routes.
The largest variation of shortest routes is in the range of mileage from 50 km to 100 km, this is because most of the locations of coal mines are relatively close to one of the ports.
Figure 1.
Coal transport modelling flow chart INERSIA.
Vol.
No.
May 2025 Nurman Nugroho, et.
Number of origin Number of origin Figure 2.
Coal transport modelling flow chart Talang Duku Tenam Tebing Tinggi Talang Duku Tenam Tebing Tinggi Travel time .
Distance .
Figure 3.
Travel distance of truck by public road .
ainy seaso.
Figure 4.
Travel time of truck by public road .
ainy seaso.
As shown in Figure 4, most of the trips with the shortest travel time are routes to Tenam Port.
A total of 68 out of 100 routes travelled with a travel time between 0-10 hours.
A total of 15 out of 100 routes travelled to Tebing Tinggi special terminal with a travel time of 0-6 hours.
Another five routes went to Talang Duku Port with a travel time between 0-2 hours.
A total of 55 mining areas are closer to Talang Duku Port, while the other 44 areas are closer to the terminal in Tebing Tinggi during the dry season.
The travelling distance in Figure 5 and travelling time in Figure 6 by the shortest route to Talang Duku Port is mostly between 100 km and 200 km and the travelling time is 4 hours to 8 hours.
The routes with terminal destinations in Tebing Tinggi are mostly spread over distances of 100-150 km and 200-250 km and the travel times are spread over the ranges of 4-6 hours and 8-10 hours.
INERSIA.
Vol.
No.
May 2025 Number of origin Number of origin Nurman Nugroho, et.
Talang Duku Tenam Talang Duku Tebing Tinggi Distance .
Distance .
Figure 5.
Travel distance of truck by public road .
ry seaso.
Figure 7.
Travel distance of truck by mine road .
ainy seaso.
Number of origin Number of origin Talang Duku Talang Duku Tebing Tinggi Tenam Travel time .
Travel time .
Figure 8.
Travel time of truck by mine road .
ainy seaso.
Figure 6.
Travel time of truck by public road .
ry seaso.
As Figure shows.
Tenam Port is the closest to 33 mine sites that are directly connected to the PT Inti Tirta Special Road with a distance between 0 km and 100 km.
A total of 18 mine sites are adjacent to the PT Sinar Anugerah Sukses Dedicated Terminal with a route distance of 0 km to 100 km, while 4 other mine sites are adjacent to Tenam Port which is connected to the PT Putra Bulian Properti Dedicated Road with a route distance of 0 km to 50 km.
As illustrated in the travel time distribution in Figure 10, most routes were travelled for 4 hours to 6 hours.
The average journey time for trucks travelling on special roads 2 hours.
Based on the Figure 8, the transport of coal by truck through mine roads during the rainy season, obtained 32 trip origins that go to Tenam Port with a travel time range of 0-2 hours.
There are 11 trip origins to the port in the 2-4 hours travel time range and there are 7 trip origins that require travel time from 4 to 6 hours.
The travel distances generated from the origin-destination analysis show that most of the routes with the destination of Talang Duku Port are between 100 km and 150 km and the average distance of all routes is 109 km.
The distribution of truck travel distances via special roads during the dry season can be seen in Figure 9.
Number of origin Coal transport by trucks over mine road Talang Duku Tenam Distance .
Figure 9.
Travel distance of truck by mine road .
ry seaso.
INERSIA.
Vol.
No.
May 2025 Nurman Nugroho, et.
Number of origin Number of origin Talang Duku Talang Duku Tenam Tenam Tebing Tinggi Travel time .
Travel time .
Figure 10.
Travel time of truck by mine road .
ry seaso.
Figure 11.
Travel time of barges .
ainy seaso.
The Batanghari River can serve 74 mines sites with a maximum reach radius of 30 km.
Four mine sites have good accessibility to the Pangabuan River.
Meanwhile, 22 mine sites have a radius of reach to the river flow of more than 30 km.
So that these coal mining areas are considered unserved by river flow during the rainy season.
The distribution graph of mileage during the rainy season shown in Figure 12 that the route travelled by barge with the destination of Tenam Port is 0-300 km.
The barge trip with the destination of Talang Duku Port was 0-50 km.
Meanwhile the barge trip to the terminal in Tebing Tinggi travelled 0-150 km.
The distance travelled by barges during the dry season with the destination of Talang Duku Port is 0-50 km.
To get to the terminal in Tebing Tinggi, the barge travelled 0-150 Meanwhile.
Tenam Port cannot serve barge transport.
The distribution of the distance of transport trips by barge can be seen in Figure 13.
The time required for the barge to travel to Talang Duku Harbour is 0-2 hours.
Meanwhile, the route to the terminal in Tebing Tinggi takes between 010 hours.
The barge travel time is summarised in Figure Number of origin Coal transport by barge Routes to Tenam Port require 0-25 hours of travel time.
total of 48 out of 70 routes from the mine to Tenam Port require 5 hours to 15 hours of travel time see in Figure.
Tenam Talang Duku Tebing Tinggi Number of origin Distance .
Figure 13.
Travel distance of barge .
ry seaso.
Talang Duku Tenam Tebing Tinggi Number of origin Distance .
Figure 12.
Travel distance of barge .
ainy seaso.
Tenam Talang Duku Tebing Tinggi Travel time .
Figure 14.
Travel time of barge .
ry seaso.
Nurman Nugroho, et.
INERSIA.
Vol.
No.
May 2025 Coal transport by railway As a result of the spatial analysis, the largest distribution of trips during the rainy season was trips to Tenam Port, 70 out of 99 routes.
There were 9 trips to Talang Duku Port and 18 trips to the terminal in Tebing Tinggi (Figure .
The distance travelled from the mining area to Talang Duku Port by train is 0-50 km.
The distance travelled to Tenam Port is between 0-200 km.
While the terminal destination in Tebing Tinggi is 0-100 km from the mining The distribution of mileage can be seen in Figure 15.
The results of spatial analysis, during the dry season, coal transport train trips with the destination of Talang Duku Port totalled 61 routes.
The trips with the destination of the terminal in Tebing Tinggi totalled 38 routes.
This means that the total number of routes during the wet and dry seasons is the same, namely 99 routes.
3 Generalised Cost The generalised cost, as shown in Figure 19, is the sum of the costs of travelling by the selected modes based on each trip origin.
In determining mode choice preferences, the generalised cost is separated into the generalised cost of multimodal transport.
4 Mode Choice Utilities The results of multinomial regression analysis on coal transport mode choice using SPSS Statistic 27 software show that the cost or tariff of coal transport trips per unit distance has no influence on the utility of mode choice.
This is because the freight transport model is inelastic to the cost or fare per distance.
However, the mode choice for freight transport is more elastic to the generalised cost in accordance with the statement in the previous literature .
Number of origin Number of origin The graph of the distribution of railway mileage during the dry season in Figure 17 shows that journeys from all mine sites to final destinations range from 0 km to 250 km.
The average distance travelled to the Port of Talang duku is 101 km, while the average distance travelled to the Terminal in Tebing Tinggi is 127 km.
The distribution of train travel time during the dry season shown in Figure 18 shows that the travel time ranges between 0-4 hours.
The average travel time to Talang Duku Port is 1.
5 hours and to Tebing Tinggi is 1.
8 hours.
Talang Duku Tenam Tebing Tinggi Talang Duku Tenam Tebing Tinggi Distance .
Travel time .
Figure 16.
Travel time of train .
ainy seaso.
Number of origin Number of origin Figure 15.
Travel distance of train .
ainy seaso.
Talang Duku Tebing Tinggi 100- 150- 200- 250150 200 250 300 Talang Duku Tebing Tinggi Travel time .
Distance .
Figure 17.
Travel distance of train .
ry seaso.
Figure 18.
Travel time of train .
ry seaso.
INERSIA.
Vol.
No.
May 2025 Nurman Nugroho, et.
Generalised cost (R.
x 1000 Parameter estimates for multimodal mode choice with the comparison of railway presented in Table 5 can be described as follows: the odds ratio value of the travel time variable on the preference for truck shows, every increase in the travel time variable for 1 hour and other variables remain constant, the possibility of choosing the truck mode as coal transport is 0.
984 times lower.
While the odds ratio value on the climate variable, if it is 1 or during rainy conditions, the probability of choosing the truck mode of transport increases 5.
046 times if other variables are Table 5.
Estimated parameters of mode choice Std.
Sig.
Exp(B) Error Preferences Truck Intercept Time Cost Climate Intercept Time Cost Climate Barge Rainy season Penghujan Dry season Kemarau Trip Origins Figure 19.
Generalised cost each origin In barge preferences, an increase of 1 hour in the travel time variable is likely to increase the probability of choosing the barge mode by 1.
177 times.
The odds ratio value on the climate variable shows that probability of choosing barge increases 3.
068 times if the rainy season and other variables are constant.
Thus, the utility of truck mode .
cOycN ) can be written as follows:
Truck Barge Railway ycOycN = Oe1.
590 Oe 0.
cNycNycN ) 1.
Where ycNycNycN is the travelling time of the truck.
Utility barge mode can be written as follows:
Figure 201.
Coal mode choice probability .
ainy seaso.
ycOyaA = Oe1.
cNycNyaA ) 1.
3% 4% Where ycNycNyaA is the travelling time of the barge.
Truck 5 Probability of Mode Choice Barge Based on the utilities that have been developed, the likelihood of the optimal mode chosen for coal transport in Jambi Province can be calculated.
Figure 20 shows that the likelihood of barge mode being selected as coal transport is higher than other modes during the rainy season.
Figure 21 shows that during the dry season, the coal shipment by railway is chosen higher than other modes.
Railway Figure 21.
Coal mode choice probability .
ry seaso.
Nurman Nugroho, et.
INERSIA.
Vol.
No.
May 2025 Based on the probability of mode choice, barge transport is the mode with the highest probability to be selected.
This indicates that the costs incurred by shippers for coal transport using barges are lower than other modes.
However, coal shipment using barges is highly dependent on climatic conditions that affect the condition of the shipping channel.
When barges cannot operate, in this case during the dry season, the cost of transporting coal using railway is lower when compared to trucks.
Referring to the results of the study, both from the generalised cost and mode choice, coal shipment by barge and railway is very suitable to be selected by mining enterprises as a distribution mode.
Conclusion After modelling the transportation of coal transport in the administrative area of Jambi Province, it can be summarised into several important things as follows.
The portion of coal shipment by truck and railway during the rainy season is almost the same, which means that both modes have almost the same transport costs during the rainy season.
Based on the proportional choice of multimodal transport during the rainy season, the probability of transport by barge via river is higher than other modes.
While during the dry season, the probability of transporting by railway is higher.
With the high probability of barging via river route, it can be stated that the most efficient transport is done during the rainy season.
Compared to transporting coal using trucks under current conditions, coal mining companies can save on transport costs when using barges and trains.
References