Optimizing Supply Chain Management in Coal Power Generation

Muhamad Iqbal Felani, Ariyana Dwiputra Nugraha and Mujammil Rahmanta

PT PLN (Persero) Research Institute, Duren Tiga Street 102, Jakarta, Indonesia

Keywords: Supply Chain Management, Optimization, Simulation, Coal Power Generation.

Abstract: Indonesian government launched Fast Track Program Phase-1 in 2009 to increase national electricity ratio

by installing 35 coal power generations with total capacity 10,000 Mega Watt. However only 25 coal power

generations had been installed by now, spread all over Indonesia. Coal necessities were supplied by 14

domestic coal mining companies. There are two factors that affect the price of coal i.e : distance and unit

price. Distance between supplier and coal power generation would determine the transportation cost while

unit price would determine the price of procurement. The aim of this research is to minimize total price of

coal by optimizing the distance and unit price (USD/Ton), allocating the coal necessities and scheduling the

delivery. The optimization would be simulated using software What’sBest. By this simulation, 24 power

plants were suggested to change their existing suppliers, while only one power plant was fitted. This change

could reduce USD 27 Million/year for total price of coal.

1 INTRODUCTION

One of the important factor in increasing the

national economic development is the availability of

low cost energy (Wang, Feng and Tverberg, 2013).

Therefore every country tries to minimize the fuel

cost in electricity supply. As a development country,

Indonesia is in progress to install coal fire power

plants with lower cost. The coal power plants spread

all over the country, while coal suppliers are

concentrated in particular island. The price of coal

consists of two factors: transportation cost

(USD/nautical mile) and unit price (USD/ton). So

the transportation factor will dominate the cost of

coal procurement since Indonesia is an archipelago

country (PT PLN (Persero) Coal Procurement

Division, 2013).

Figure 1: Map of Coal Power Plant and Coal Supplier

Spread in Indonesia.

Every power plant requires a specific coal that

must be supplied by appropriate supplier. In

consequence of this problem, the utility should

arrange the best supply chain management to

minimize the cost. In case utility could minimize

the distance between power plant and appropriate

supplier, the cost would be minimal.

There are 25 power plants with specific coal

requirement and 14 coal supplier that would be

optimized in this paper. The purpose of this paper is

to optimize supply chain management by

minimizing the distance between power plant and

appropriate supplier and ensure that production

capacity of supplier could serve the power plant as

long as contract duration. Software What’sBest is

used to simulate this problem. The results of the

simulation are the demand – supply mapping,

allocating the coal requirement and scheduling of

shipping.

2 METHOD

This paper simulated the coal supply chain

management planning according to the real

condition. The simulation would be processed by

What’sBest software. The first step in this research

was data collection. Data used in this simulation;

 Information about coal power plant as a demand,

involving the name of Power Plant, the location,

460

Felani M., Nugraha A. and Rahmanta M.

Optimizing Supply Chain Management in Coal Power Generation.

DOI: 10.5220/0006250904600463

In Proceedings of the 6th International Conference on Operations Research and Enterprise Systems (ICORES 2017), pages 460-463

ISBN: 978-989-758-218-9

the quantity of coal demand, the quality

requirement of power plant, safety stock level,

and jetty capacity.

 Information about coal supplier, involving the

name of supplier, the quality of coal, unit price

(USD/ton), transportation cost (USD/Nautical

miles), productivity (ton/year), contract duration

and ship capacity.

 Information about ocean condition in Indonesia

yearly, involving the height of wave, the velocity

and direction of wind (Fig.2).

 Distance and lead time between supplier and

demand. The distance is determined by

Geographic Information System (GIS)

Directorate General of Marine Transportation,

The Indonesian Ministry of Transportation.

Therefore the distance is not straight line, but

following the cruise lane (fig.3).

Figure 2: Wind and Wave Ocean Condition in Indonesian.

Based on Fig.2 above, the information is

converted in binary number. Number 1 means a high

wave that affect the ship velocity, and number 0

means normal condition that not affect the ship

velocity.

Figure 3: Wind and Wave Ocean Condition in Indonesian.

The velocity of ship in Fig.2 and the distance

between supplier and demand in Fig. 3 results in

determining lead time.

All of data above were processed into a database.

The next steps were allocating the coal necessity and

scheduling. The steps in simulation process could be

seen in Fig.4 as follow:

Figure 4: Steps of Simulation Process.

Some decisive variables must be considered

when allocating the coal demand. These variables

are presented by mathematical equation;

 Supplier i only could send coal q to the power

plant j as if coal q has the same specification

with power plant requirement. In other word, the

caloric value of the coal from supplier must be

appropriate with caloric value required by power

plant. This decisive variable is capable or not,

noted by Y

i,j

 Coal q ϵ Q was supplied by supplier i ϵ I to

power plant j ϵ J. This decisive variable is the

amount of each type of coal which is delivered

by supplier, noted by X

i,j,q

The purpose of this allocation is to minimize total

cost (f) of coal. There were 2 main components

contributing in total cost; procurement cost (f

) and

transportation cost (f

). Therefore this function can

be determined as follows:

Minimize f = f1 + f2 (1)

Procurement cost is multiplication between unit

price (P

i,q

)of coal q to supplier i with total amount

i,j,q

) of coal q that is delivered by supplier i to

power plant j. Therefore the formula of procurement

cost (f

) can be determined by:

,,ijq

iI jJ qJ

 





,iq

(2)

While transportation cost is multiplication between

transportation cost (TC

i,j

) from supplier i to power

plant j with total amount (Xi,j,q) of coal q that is

delivered by supplier i to power plant j with distance

) from supplier i to power plant j. Therefore the

formula of transportation cost (f

) can be determined

by:

,,ijq

qQ iI jJ







,ij

(3)

To achieve the optimal function above, then

allocation model must comply with these limitation:

1. Maximum amount (X

i,j,q

) of coal q that could be

procured by power plant j is limited by

Optimizing Supply Chain Management in Coal Power Generation

461

maximum capacity (O

i,q

) of coal q which is

available in supplier i comply with contract

document between supplier and utility, this

limitation could be formulated as:

,, ,ijq iq







∀i ϵ I, j ϵ J, q ϵ Q

(4)

2. Every power plant has its typical caloric value

of coal. Delivery of coal could be processed as

if caloric value of coal in supplier comply with

power plant‘s caloric value. The coal which has

caloric value beyond the range could be

processed. This limitation could be formulated

as:

(5)

3. The coal q from supplier must be able to fulfil

the demand (D

) every power plant

. This

limitation could be formulated as:

,, ,ijq jq







∀j ϵ J, q ϵ Q

(6)

The final step of the simulation is scheduling,

which time of delivery of coal would be determined.

The limitation in scheduling is that power plant j

only could receive of coal from supplier i once a day

(t). This limitation can be formulated as RC

i,j,q,t

3 RESULT

The simulation gave a result that from total 25

existing power plant, there are 24 power plants

should change their supplier due to cost optimizing.

This simulation also gave the optimal amount of

coal that should be procured by power plant, it calls

optimal allocation. Therefore all of power plants

could make a coal procurement plan effectively. The

next step is scheduling. All of the limitations of

allocating and scheduling are conducted by

What’sBest software. The scheduling covers all

information about when supplier must deliver their

coal to power plant, how much coal that must be

delivered to the power plant, when the coal would be

received by power plant considering the ocean

condition and how much coal that available in power

plant inventory as consequence of lead time

variance. The example of scheduling table could be

seen in Table 1.

From Table 1 above could be seen that power

plant “A” would be supplied by supplier “1” as

much as 7,500 ton in September 24

(purple cell)

and would be received by power plant “A” in

September 29

(yellow cell). Safety stock level in

power plant “A” in September 24

is 54,080 ton

(orange cell). This safety stock level would be

maintained in 25 operating days. Lead time is

presented as green and blue cell. Green cell is for

normal condition (weather) while the blue one is for

bad condition (weather).

4 DISCUSSION

Total cost (procurement cost and transportation cost)

before simulation and after simulation was

compared to evaluate it significance. Transportation

cost before simulation was not available due to poor

of information. There was no data about the amount

of coal that is delivered by supplier to power plant.

While total cost after simulation is determined by

model. The comparison before and after simulation

could be seen in Table 2.

The Table 2 above informed that after simulation

procurement cost decreased as much as

24,110,173.53 USD per year. In case it is assumed

that amount of coal that is delivered by supplier to

power plant before simulation is equal to after

simulation, then it definitely results that

transportation cost after simulation is less than

before simulation. In other word we could say that

there is a benefit after simulation.

Table 1: The Example of Scheduling.

Power

Plant

Supplier

Total of

coal

Ship

capacity

Initial

Capacity

September

Normal* Bad* 24 25 26 27 28 29 30

Level Inventory 51,750 54,080 51,990 57,420 55,350 57,420 58,710 56,640

1 755,550 7,500 Received 0 0 7500 0 0 7500 0

Lead Time 5 6 Order 7500 7500

Level Inventory 47,040 48,652 47,771 50,889 49,008 50,889 50,244 48,363

2 552,151 5,000 Received 5,000 0 0 5,000 0

Lead Time 16 24 Order 5,000 5,000

3 134,632 7,500 Received 0 0 0 0 0

Lead Time 2 2 Order

ICORES 2017 - 6th International Conference on Operations Research and Enterprise Systems

462

Table 2: The Comparison of Total Cost.

Cost Component

Before Simulation

(USD)

After Simulation

(USD)

Benefit

(USD)

Procurement Cost 897,183,961 873,073,787.5

24,110,173.5

Transportation Cost Not Available 363,454,502.5

Not Available

Total Cost 897,183,961 + NA 1,236,528,290

Last year, Indonesian government launched

instalment of power generation program with total

capacity 35,000 Mega Watt (MW). This program

will be completed in 2019. Majority of them are coal

fired power plant. The most critical problem in coal

power generation is the reliability of coal supply. It

will need a huge demand of coal in 2019. Therefore

local coal suppliers need to be mapped. Then

scheduling each power generation will be arranged

and optimized. The method in this research is being

used to solve the problem.

5 CONCLUSIONS

The research proved that the model simulation could

optimize total cost of coal demand in power plant.

There are 24 of 25 existing power plant that should

change their allocating and scheduling planning to

increase cost effective. By optimizing supply chain

management it could reduce total cost at least

24,110,173.53 USD per year. The method in this

research is being used to ensure the reliability of

coal supply to support government’s program;

Instalment of 35,000 MW power generation.

ACKNOWLEDGEMENTS

The author thanks the PT PLN (Persero) Research

Institute for the chance and support. The author also

thanks to PT PLN (Persero) Head Quarter Division

of Coal Procurement for valuable data and direction.

Thanks to Indonesian Coal Fire Power Plant for

group discussion, data and valuable supports during

this research.

REFERENCES

Moore, R., Lopes, J., 1999. Paper templates. In

TEMPLATE’06, 1st International Conference on

Template Production. SCITEPRESS.

Smith, J., 1998. The book, The publishing company.

London, 2

edition.

J. Wang, L. Feng, and G. E. Tverberg, 2013. An analysis

of China’s coal supply and its impact on China's

future economic growth,. Energy Policy, vol. 57, pp.

542–551,

PT PLN (Persero) Head Quarter Division of Coal

Procurement,2013. Supply and Demand Coal Data,

Jakarta.

Indonesian Agency for Meteorology, Climatology and

Geophysics, 2014. Indonesian Monthly Ocean

Condition, Jakarta..

Indonesian Ministry of Transportation, 2015. Geography

Information System, Jakarta.

Optimizing Supply Chain Management in Coal Power Generation

463