2008 Workshop on Power Electronics and Intelligent Transportation System

Simulation and Optimization of the Power Station Coal-Fired Logistics System

Based on Witness Simulation Software
Yabin Li 1, Rong Li 2
School of Mechanical Engineering,
North China Electric Power University, Baoding , HeBei, 071003, China
2
Information and Network Management Center,
North China Electric Power University, Baoding , HeBei, 071003, China
lybhnjz@163.com
1

stockpile optimization strategy, and established the

dynamic project model of order time and order quantity.
Literature [5] has studied an integrated coal transportation
and stockpile model under the condition of rail direct
transportation.
On the one hand, about the above-mentioned
documents conditions are limitedly considered, and the
study of human resources is also lack in thermal power
station coal-fired transportation and stockpile problems;
on the other hand, it is crucial not to realize the simulation
of the power station coal-fired logistics system, so it is
very difficult to find what time and what position
bottlenecks occur in the power station coal-fired logistics
system and thus it is absent to promptly and effectively
improve the characteristics of the power station coal-fired
logistics system.
Therefore, this paper, according to the premise of coal
boiler requirements, on the basis of the comprehensive
analysis of coal-fired rail transportation and stockpile that
are two important links of logistics system, major studies,
sets up and simulates by Witness2006 software the
multi-cycle logistics system of coal-fired thermal power
stations. And by demonstrating the entire process of its
logistics systems, the model makes users clearly
understand its logistics problems in the process and adjust
the model parameters in a timely. And the model makes
the total costs of the logistics system as the optimal goal,
which includes coal purchase cost, transportation cost,
stockpile cost, capital cost, and applies Witness software
optimization modules for optimization, so as to implement
the best value of the power station coal-fired logistics
system.

Abstract
Based on analyzing characteristics and relations of the
power station coal-fired logistics system, including the
choice of coal suppliers, the railroad transport system, the
power station stockpile system, the power station coal
transfer system and human resources, by Witness software
the thesis sets up a simulation and optimization model of
fuel coal logistics system and detailedly analyzes one
example. The model results not only macroscopically
image bottleneck points and resources availability, but
define optima separate stage ordering goods batches and
stockpile quantities. The thesis is powerful and significant
for thermal power station to establish scientific
reasonable logistics system strategies, reduce the cost and
increase competition ability.

1. Introduction
According to statistics, the velocity of national electric
power increase is very quick in the last few years. In the
total output of electrical energy, the thermal power
occupies about 80%. Among them, the cost of a thermal
power station coal-fired logistics system in China,
including the purchase of coal, transport and storage et al..,
is often accounted for 60% to 70% of the cost of power
generation. Under the condition of guarantying coal-fired
supply, how to cause the least coal-fired logistics system
expense, is the primary mission and goal of the coal-fired
logistics system administration [1].
In recent years there are also some literatures on
transport and stockpile models of coal-fired thermal power
stations. Literature [2] has theoretically merely studied the
power station coal-fired logistics system's characteristic,
and not given the substantive solution. Literature [3] has
studied coal purchase and transportation questions on
Taiwan Electricity companies, and established the
transportation plan of many suppliers selection and the
stockpile control mix integer project model. Literature
[4] has studied the thermal power station coal-fired

978-0-7695-3342-1/08 $25.00 2008 IEEE

DOI 10.1109/PEITS.2008.103

2. Analysis of the power station coal-fired

logistics system
The power station coal-fired logistics system has three
major components, namely, suppliers selection, railroad
transport systems and power stations.

394

coals arriving to factories are affective to coal-fired

qualities and quantities [8].

2.1. Suppliers selection

2.1.1. Guarantying coal supplies. When coal production
capacity or supply bottlenecks encounters blocked, it is
easy for the supply chain to trigger coal-shortage crisis.
Therefore, power enterprises can unite with coal
enterprises to handle an ore for reaching the purpose
participating in competition on the headstream, and also
can sign the middle-long supply and demand contract with
coal enterprises under strategic partnership, which can cut
down incomplete coal risk, coal purchase cost, and build
healthy developed economic supply chaining [6].

3. Modeling principle by Witness

Based on the thermal power station coal-fired logistics
system, the modeling principle is: power stations may
select suppliers according to suppliers' quote, freight, coal
composition and delayed payment, et al.; the quantities
and the time of supplying fired coals depend on power
stations demands and the minimum cost principle;
coals transport systems should be consistent with the
requirements of railway system and ensure the economy
transport weigh and transport efficiency; power station
stockpile system should ensure that the optimal stockpile
costs, and the use of human resources, as well as the
scheduling of electrical equipments should meet maximize
efficiency.

2.1.2. Ensuring coal qualities. Because the fired coal has

the strict request to the indexes, when power stations
select suppliers they must consider qualities and quantities
of coals in the lowest purchase cost, as well as after mixed,
to be able to achieve the coal-fired requests.

2.2. Railway transport systems

4. Constructing models

More than 80 % of power station coals need railway

transports, and the vast majority of coal rail transports are
used to rail direct transport. Of course, this way has
certain requirements on the number of direct cargo trains,
the consignor loading capacities, the rail carrier capacities,
economic freight weights, plant unloading capacities and
stockpile system conditions.

4.1. Parameters definition

The model input parameters: Q(k) as the demand for
power station at stage k, t; E as the proportion of a
supplier supplies occupying total demands (top limit),%;
Ci(k) as order cost of coal i at stage k ,yuan/time; Bi(k) as
unit railway direct transport costs of coal i at stage k,
yuan/t; Pi(k) as unit purchase price of coal i at stage k,
yuan/t; G(k) as power station unit storage costs at stage k,
yuan/t; S as stockpile capacity of a power station, t; Li as
lower limit of railway direct transport capacity, t; S(k) as
safety stockpile quantities of a power station at stage k, t ;
Ui as the arriving rate of supplier i providing coals ,%; U
as the minimum acceptable arriving rate of a power
station,% ;R as the discount rate,%; Ti as the delayed
period of a power station paying for coal i, d; Xij as index j
value of coal i,%; Yj as a boiler requesting coal index j
value,%; W(k) as the wages of power station workers at
stage k, yuan/month.
The model output parameters: Zi(k)as the quantities of
ordering coals from suppliers i at stage k, t, V(k) as the
stockpile of power station fired coals at initial stage k, t;
Costi(k) as power station coals purchase total costs and
transport total costs from supplier i at stage k; Stock(k) as
power station stockpile costs at stage k; Cost as the total
cost of thermal power station coal-fired logistics system.

2.3. Power stations

2.3.1. stockpile systems. In the coal-fired logistics system,
the plant stockpile system is essential aspect, which is
closely related to security, stability and economic
operation. It is decided by the factors that are power
station demanding fuel quantities per day, security
stockpile, ordering cycle, delayed arrival time, the ability
unloading capacities and so on. [7].
2.3.2. Factory coal-fired conveying systems. It is mainly
composed of dropping coal machines, belt conveyers,
screening machines, measurement machines, sampling
machines, transportation machinery and so on. This
system mainly researches unloading coals machinery
related to unloading ability and transport machinery
related to the boiler load.
2.3.3. Human resources. In this study of the logistics
system, the human resources mainly are for checking and
unloading coals, machinery service and conveying coals.
And, it is very important for quality testers to check coals
qualities and quantities, which not only relates the round
turn economic interest, but also relates the accuracy of
coal-fired consumption computation used for electricity
generation. And the magnitudes of moisture contents of

4.2. Assumption of constraint conditions and

determination of objective functions
Construction of the model based on the following
assumptions:
(1) In order to maintain consistent with the ordering

395

fired coal plan, assume the research cycle of the model for
one year, namely, designing a one-year coal-fired
subscription plan. And the research phases of the model
are monthly divided into 12 stages, and the specific
research object is one day;
(2) In order to ensure the security and stability of
coal-fired supplies, the level of a power station depending
on each supplier is not more than 50%, and each supplier
provide only one type of coal, namely:
Z i ( k ) 0 . 5 Q ( k ) ,i=1,2,,nk=1,2,,12 (1)

The objective function is set to the minimum total cost

of the logistics system, namely:

(3) Assuming that applying the rail direct transport

mode, trains have minimum and maximum capacity
constraints;
(4) In order to the actual situation of unequal paths from
suppliers to power stations, and to ensure the accessibility
of the model, the model utilize different transport costs per
km to express unequal paths, namely, under the conditions
of different fired coals with unequal transport costs per km
and the coal-fired invariable unit transport cost per km at
every stage , the paths in the model show the equal
distance from different suppliers to the same power
station;
(5) In general, the coal enterprises pay for the freight in
advance, and after coals arrive to the plant the power
station together pays for purchase costs and the freight.
Therefore, in order to reflect capital time value on the
impact of the coal-fired cost, the model introduce a
discount rate and a power station delayed payment period
for coal i to accurately reflect the true value cost, and the
delayed period is different with different suppliers;
(6) Ordered coals arrive in the same stage [9];
(7) The stockpile is the quantities of the beginning of
each stage (month, day), and unit storage cost is related to
storage quantities and stages, and there are the largest
stockpile and safety stock restrictions, namely:

Total cost of the logistics system consists of three parts,

namely, coal-fired purchase and transportation costs,
stockpile costs and labor costs.

(k ) + V (k ) S ,

i, k

(2)

i =1

V (k + 1) = V (k ) + Zi (k) Q(k) S (k) , i, k

(3)

i =1

(8) The coals-mixed quality requirements of the boiler

mainly reflect coal-fired caloric power, the volatile
content, moisture content, sulfur content and ash content,
and so on. These indexes must be controlled to set limits.
When the raw coal arrives at the power station, there is the
lowest request regarding the coal arriving percentage.
After quality testers checkout indexes and quantities, it
can be arranged by the dropping coal personnel to unload
the coal to the coal field, namely:
n

ij

Z i (k ) ()Y j Q(k ) i , j , k

(4)

i =1

U
i =1

Z i (k ) U Q(k ) , i , k

(5)

Costi(k) = Ci(k) + Z i (k) (Pi(k) + Bi(k))/(1+ R)Ti

(6)

Stock(k)=G (k) ( Z i (k) +V(k+1)+V (k))/2 (7)

i=1

12

Cost = [ Costi (k) + Stock(k) + labors W(k) ]

k =1 i =1

i, k (8)

4.3. Construction of simulation and optimization

models

Figure 1. Simulation and optimization models of the

power station coal-fired logistics system

Based on assumptions, applying Witness software

constructs the model as shown in Figure 1. In the Model,
the suppliers may have more than one, and there is a
coal-loading machine and a loading buffer zone
corresponding to one supplier. Under assumptions (4),
corresponding to the power station there are a number of
trains and one circular direct railway (track 1 and track 2).
The model also orderly sets up the power station buffer to
unload coals, three quality testers, two coal-whippers, the
power station stock, one dig and transport coal machine
and nine operators, as well as one boiler and so on.
The coal supplies in the model depend on the needs of
power stations, with ton as coal units and hours as
simulation time; the adjusted cycle of coal-unloading and
coal-mining machines are one time per 100,000 time
work, and the adjusted time is 24-hour long. And
maintenance cycle meets NEGEXP (20,1) distribution,
and repairing time meets POISSON (12,2) distribution; a
worker works eight hours under the condition of three
shifts of work system, and so on.
The model applies optimizer 4.3 module for
optimization. The model makes assumption conditions as
restrictions, cost as the objective function and Simulated
Annealing method as algorithms.

5. Case study
5.1. Parameters assignment
Aimed directly at South China thermal power station A
coal-fired logistics condition, this paper applies the model

396

After the system operation, the simulation results can

macroscopically show the bottlenecks of the logistics
system operation and the information of resource
utilization efficiency at all aspects, which can be used for
targetedly repeatedly adjusting logistics system resource
allocation to ensure the final maximize resource utilization,
and thereby minimize logistics costs. After the
optimization, optimization results of the procurement
strategies, the stock quantities and the total cost as shown
in table 2.

of simulation and optimization. The plant purchase coals

from three coal suppliers, with the procurement period of
12 months and one month as a stage( k =12).
Initially stockpile V(0) = 105000 t, E=0.5, Ci (k) = 900
Yuan, G(k) = 0.4 Yuan, S = 420000 t, Li =1200t,
S(k)=80000 t, R=0.02, Ti=10 days, U1=0.98, U2=0.97,
U3=0.97, U=0.97, W(k)=5,000 Yuan, 3 quantity testers, 9
operators and other variables as shown in table 1.

5.2 Analysis

Table 1. Parameters of the power station coal-fired logistics system

k

Time

B1

1
2
3
4
5
6
7
8
9
10
11
12

0-720
720-1440
1440-2160
2160-2880
2880-3600
3600-4320
4320-5040
5040-5760
5760-6480
6480-7200
7200-7920
7920-8640

91.35
91.35
89.25
89.25
87.15
89.25
89.25
89.25
86.1
86.1
90.3
91.35

B2

B3

P1

P2

P3

96.6
96.6
94.5
94.5
92.4
94.5
94.5
94.5
90.3
90.3
95.55
96.6

94.5
94.5
92.4
92.4
90.3
92.4
92.4
92.4
89.25
89.25
93.45
94.5

388.5
367.5
367.5
378
388.5
399
399
388.5
378
367.5
378
388.5

393.75
362.25
372.75
378
393.75
399
399
393.75
378
372.75
378
388.5

393.75
367.5
378
383.25
393.75
399
393.75
388.5
383.25
372.75
378
388.5

202860
210168
273168
275436
277200
293832
315000
357084
319284
268884
252000
231084

X1

X2

X3

1*
2*

0.25
0.02

0.35
0.03

0.3
0.04

0.3
0.03

Note:1*-volatilize content 2*-sulfur content

Table 2. Optimized results of the power station coal-fired logistics system
k
1
2
3
4
5
6
7
8
9
10
11
12

Z1k
101430
105084
136584
137718
138600
0
115332
178542
159642
134442
126000
0

Z2k
0
105084
136584
137718
0
0
0
0
159642
134442
96642
0

Z3k
80430
105084
136584
94332
138600
0
157500
178542
0
134442
126000
0

Vk
84000
189084
325668
420000
420000
126168
84000
84000
84000
218442
315084
84000

Cost(Million)

1200.293

transport capacity constraints, transport quality constraints,

stockpile capacity constraints, security stock constraints,
boiler requirements for coal qualities, as well as blending
ratio, et al.. Through simulation and optimization of the
model, optima separate stage ordering goods batches and
stockpile quantities can be defined, which provide
fundamental strategies to formulate scientific and rational
logistics system. The analysis of the practical sample
through the model further explains the construction
process of the model and verifies the validity and the
feasibility of the model.

6. Conclusions
Based on the analysis of characteristics of the thermal
power station coal-fired logistics system, applying witness
simulation software builds the thermal power station
coal-fired multi-cycle logistics system model under the
condition of a rail direct transport. Objectives is the
minimum total cost of the logistics system , including coal
purchase costs, transportation costs, stockpile costs,
capital costs. Logistics system take full account of the
various constraints relying on the needs of a coal-fired
power station, including suppliers selection, railway

Acknowledgement

397

It is a project supported by North China Electric

Power University scientific research fund.

References
[1] Hyde M(1998). Optimizing plant and system coal
inventories. IEEE Transaction on Power Systems, Vol. 32,
No. 1, pp. 337-342
[2] Ying Li, and Mengjun Li(2007). Study on logistics cost
control of thermal power plant. Enterprise Logistics, Vol.
26, No. 5, pp. 117-118
[3] Shih Li-Hsing(1997). Planning of fuel coal imports using
a mixed integer programming methodIntJProduction
Eeonernics, Vol. 51, No. 6, pp. 243-249
[4] LF Li, and PQ Huang(2003)Research of fue1 inventory
management in power plantIndustrial Engineering and
Management Vol. 32, No. 6, pp. 68-71
[5] Xueming Cao, and Boliang Lin(2006). An integrated coal
transportati0n and inventory model under condition of rail
direct transportation. Journal of Beijing Jiaotong
University, Vol. 30, No. 6, pp. 27-31
[6] Haixian Yang, and Xinwei Li(2004). Improvement
electricity coal supply chain. China Power Enterprise
Management Vol. 21, No. 7, pp. 30-31
[7] Lixia Chen(2006). Sate storage management of coal in
power plant. Industrial Safety and Dust Control, Vol. 32,
No. 2, pp. 61-63
[8] Xichun Fan, and Xiaohua Wang(2004). Effect of
coal-acceptance on stock coal weight in plant
coal-fired. .Jilin Electric Power, Vol. 170, No. 9, pp.
51-52
[9] Lingfeng Li, and Peiqing Huang(2003). Research of fuel
inventory management in power plant. Industrial
Engineering and Management, Vol. 25, No. 6, pp. 67-71

398