


  
TRANSACTIONS ON ENVIRONMENT AND ELECTRICAL ENGINEERING ISSN 2450-5730 Vol 1, No 3 (2016) 

© Raivo Melsas, Argo Rosin and Imre Drovtar 

  
Abstract— Demand side response enables cost optimization for 

energy systems and industrial consumers. In many countries, it is 
not widely used because of implementation complexity. One of the 
solutions for applying demand side response is industrial process 
scheduling according to the energy market needs. From the energy 
system point of view, process scheduling implies load scheduling. 
The aim of this paper is to provide a solution for load scheduling 
by implementing value stream mapping, which is a 
straightforward enough for production management. Decision 
makers in the industry should have a clear understanding about 
positive effect from load scheduling and its effect to production 
outcome and process availability. Value stream mapping is a well-
known process optimization tool from lean production philosophy. 
The aim of value stream mapping is to shorten the lead time of 
industrial processes and to reduce the intermediate stock amounts. 
By complementing value stream map with process energy intensity 
and energy stored in intermediate stocks, we can promote load 
scheduling possibilities. Our methodology provides a tool that is 
understandable and traceable for industry-minded decision 
makers. Finally, we present a real life test example for the new 
methodology, which is based on the production process of a district 
heating plant.  
 

Index Terms— demand response, energy storage, load 
management, load scheduling, value stream mapping 
 

I. INTRODUCTION 
OAD scheduling (LS) as part of demand side response 
(DSR) must meet the needs of industry. One of the effects 

for the industry appears when load consumption is shifted from 
periods of high electricity price to those of low price. As a 
result, cost savings can be achieved by means of reduced 
consumer demand in high price periods. This requires better 
production planning, which is related to production 
management. 

Previous DSR studies have resulted in the following: static 
Stackelberg game theory for voluntary load curtailment 
programs [1]; numerical calculation method for DSR when a 
battery energy storage system (BESS) is utilized [2]; solutions 
for DSR by means of automatic lighting [3]; DSR with micro-
CHP systems [4]; an overview for DSR methods in high 
 

This research was supported by the Estonian Centre of Excellence in Zero 
Energy and Resource Efficient Smart Buildings and Districts, ZEBE, grant 
2014-2020.4.01.15-0016 funded by the European Regional Development Fund. 

consumption industries and examples of market tools that 
support DSR [5]. In [6] an automated complex system for LS in 
industry is described, which takes into account stock 
restrictions, maintenance schedules, and crew management. All 
the necessary inputs are analyzed with a fuzzy/expert-based 
system combined with an optimization module. As a result, the 
system is able to identify whether and how much the industrial 
plant can participate in a DSR event. In [7] and [8] DSR is 
addressed as a part of the following main load shaping 
strategies:  

a) conservation - energy saving is achieved through 
static methods; 

b) load growth - energy consumption is increased when 
an energy system has surplus energy production; 

c) valley filling - load is increased through the off-peak 
periods or keeping stable consumption; 

d) peak clipping - energy consumption is decreased in 
peak periods; 

e) load shifting - peak consumption is shifted from peak 
periods to non-peak periods; 

f) seasonal load reduction - annual energy peaks are 
reduced. 

LS is used mainly in “load shifting” strategy e; however, in 
some cases, “valley filling” strategy can be utilized as well. To 
use the LS, an industry must have the following one or several 
DSR options [8]: cooling equipment with cooling storage, 
heating equipment with heat storage, dual fuel systems that can 
operate either on electricity or on an alternative fuel, 
discretionary loads and process equipment that can be shifted 
during a short period or material handling equipment with 
storage possibilities (silos, stock, etc.). This paper focuses on 
the last option by looking at process as a whole in order to find 
LS solutions. In addition, it provides a method applicable in 
industry for outlining the possibilities with LS as a part of DSR 
by using value stream mapping (VSM). VSM is applicable in 

R. Melsas, A. Rosin and I. Drovtar are with the Department of Electrical 
Engineering, Tallinn University of Technology, Ehitajate tee 5, 19086 Tallinn, 
Estonia (e-mail: raivomelsas@gmail.com, argo.rosin@ttu.ee, 
imre.drovtar@gmail.com). 

Value Stream Mapping for Evaluation of Load 
Scheduling Possibilities in a District Heating 

Plant 
Raivo Melsas, Argo Rosin and Imre Drovtar 

L 



various ways. Originating from Toyota Production Systems [9], 
it was further elaborated and adjusted to find solutions for 
different problems in the production process. For example, 
VSM is used to solve quality problems [10]. This paper 
elaborates on VSM. Our proposal is to use it for indicating a 
possibility to shift an electrical load from a high price period to 
a low price period and utilize an intermediate stock for energy 
storage [11], [12], [13]. This paper will provide a 
straightforward solution for the industry in order to apply LS 
effectively and which is easily applicable. 

II. II. ENERGY PRICE AS A DRIVER FOR LOAD 
SHIFTING 

LS can yield an economic effect under rational consideration. 
Cost reduction can be achieved by taking advantage of energy 
price fluctuations during a day. It is reasonable to have demand 
response implemented in countries where an electricity pool 
exists and hourly based spot prices are known for a short period 
ahead. As a result, industries can plan their production 
according to the spot price. Fig. 1 shows an average 24-hour 
electricity market spot price in Estonia in 2014 [14]. As can be 
seen, electricity spot price is typically higher from 07:00 A.M. 
to 8:00 P.M. In general, there is at least 10-euro price difference 
during a day and a night. Considering the price peak and dip 
approximately 20-euro difference per MWh exists during a day. 
On average, price difference during a day is 15-euro MWh. 

However the consideration above includes only electricity 
price; in addition, there are some fluctuations in grid service 
price as well. From 12:00 A.M. to 8:00 A.M. (11:00 P.M. to 
7:00 A.M. in winter time), the grid service price is lower. This 
period is not overlapping 100% with a low spot price period. 
We will call this period (12:00 A.M. to 8:00 A.M.) as the Low 
Price Period (LPP) and the other period during a day as the High 
Price Period (HPP). Grid tariffs depend on grid connection 
voltage level and connection capacity (amps). In the following, 
we will describe one example to examine the daily price 
difference for the industry in Estonia. 

At substantial electricity consumption, an industry is usually 
connected into a middle voltage (MV) grid. In that case, 
Estonian grid service price is 14.5 euros per MWh from 8:00 
A.M. to 24:00 A.M. and during LPP 8.3 euros per MWh [15]. 
For some customers, no time difference is applied; the price for 
grid services is constant in time- 12 euros per MWh. Fig. 1 
shows both the grid price and the spot price fluctuations. 

In 2014, an average electricity spot price for LPP was 29.5 
euros per MWh and with grid price fluctuations it amounted to 
37.8 euros per MWh, which we call as the Low Price Period 
Price (LPPP). In 2014, an average HPP spot price was 40.9 and 
together with grid tariff, the average number was 55.1 euros per 
MWh, which we call as the High Price Period Price (HPPP). 
For clarity, 55.1 is an average found on hourly bases, i.e. spot 
price + grid price. 

Based on the data provided, we can calculate potential 
savings for industry under LS. Potential cost saving is 31%, 
which is calculated by (1): 

 

.
HPPP

LPPPHPPP
CS

−
=  (1) 

 
In general, it can be concluded that grid tariff fluctuation has 

an important role in LS in Estonia, as the average difference in 
the spot price between HPP and LPP was 11.4 euros per MWh 
and grid tariff will add extra for the difference between HPPP 
and LPPP, according to [15], tariff depends on the grid 
connection parameters. 

The shape of the electricity spot price in Fig. 1 can be 
considered as a typical shape of the daily demand of electricity 
as well; a similar shape of demand can be found in various 
places, e.g., even in South Africa [16]. Thus, DSR has a positive 
impact to overall efficiency to energy systems, not only to an 
industry itself. 

 

III. IMPROVEMENT OF VALUE STREAM MAPPING 
METHODOLOGY FOR EVALUATION OF LOAD 

SCHEDULING POSSIBILITIES 

A. Load Scheduling at Intermediate Storage Use in 
Production 

In the following, LS possibilities are examined for 
intermediate storage use as an energy saving unit, to enable 
shifting of energy intense production from HPP to LPP. In LS, 
it is important to understand energy intensive production units 
and their overall role in the production. Methods from lean 
philosophy [LP] can be used here. We propose to improve the 
value stream mapping (VSM) methodology with LS principles. 
Today LP is a leading production management philosophy. 
VSM is used to plan production as efficiently as reasonably 
possible. Energy intensity can be added in a process as 
additional information for a VSM, which will give a good 
overview about the possibilities in energy saving. Fig. 2 gives 
an overall picture of a typical VSM, elaborated with process 
energy intensity and amount of energy stored in the production 
intermediate storage. 

 
Fig. 1. Average Estonian 2014 electricity spot price during a day. 

20

25

30

35

40

45

50

55

60

65

1 3 5 7 9 11 13 15 17 19 21 23

Pr
ic

e 
(E

U
R

/M
W

h)

Hours in a day

spot+grid price on HV line spot price



 
 
As VSM gives an overview of the production planning and 

can additionally give information about energy intensity, the 
possibility of the LS should be studied in detail. From VSM, we 
will know if an energy intense process is at the same time a 
bottleneck in production. If it is not true, then the conclusion is 
that this process is not 100% utilized in time and LS can be 
implemented without increasing process capacity. Alternatively 
costs for increasing process capacity (CPC) need to be 
calculated. 

Next, we focus on the storage. If production is shifted in time, 
storage volume can increase as compared to the state without 
LS utilized. Storage increase may need additional investments, 
which should also be considered as costs for storage (CS). Also, 
costs of the process (CP) itself for LS should be taken into 
account. As an example CP related to LS can be an increase in 
labor costs due to night shifts. Finally, if the costs related to LS 
are lower than a possible income from LS (ILS), the LS can be 
implemented in the industrial process. This criterion is given in 
the following (2): 

 
ILS<CP+CS+CPC . (2) 

 

B. Methodology for LS with VSM. 
The main initial goal of LP and VSM was to reduce lead time 

(L/T), i.e, the time it takes to move a produced piece all the way 
in the process or production from start until to the end. As a 
result, many costs or wastes, as defined in LP, will be reduced. 
As shown in (2), income from LS should be higher than costs 
related to it because LS can increase the intermediate stock and 
due to that it has a negative effect on the L/T as well. VSM is a 
tool that will help to find the processes which have a reasonable 
effect and income. To achieve that, cost saving potential must 
be estimated. In section II, we showed a possible gain from LS 
in Estonia. The next step is to define the energy consumption of 
the processes. It is useful to combine that with VSM. If a 
company already has VSM, then energy consumption should be 
added into the VSM. VSM will highlight important information 

to be considered for LS. Most importantly, the following 
information must be taken into account: 

a) Identify if the process is a bottleneck in production. If 
the process is a bottleneck, then load shifting is 
usually impossible without investment into the 
process output increase. 

b) Process cycle time (C/T) is slower or faster than its 
next process C/T. C/T in VSM describes how often a 
part or a product is completed by a process. If the 
process C/T is slower than the process coming next, 
it is possible to increase intermediate stock at the end 
of the process coming after rather than at the end of 
the first process. In case the first process is faster 
than the next one, the LS applied will increase the 
intermediate stock. 

c) Weather process uptime (UPTIME) is high or low, 
shows important information about process 
reliability. A low reliability process has a negative 
effect on LS. 

d) If stock between processes is high or low, the reasons 
should be found out before applying load scheduling. 

e) Change over time (C/O) is long or short. C/O is the 
time period for switching the process unit from one 
product to another. It can show also time for cold 
start of the process. C/O can highlight important 
information about the process start up time. 

The best way to start defining the energy consumption of a 
process is to make the consumer list. Consumer list should be 
based on processes described with VSM. The consumer list 
may be available in the facility electrical department. In that 
case, consumer list is usually based on the power cabinets and 
it has to be made process by process. If the consumer list is not 
available, it should be done from scratch. As the purpose is to 
find an initial energy intensive process, the consumer list can 
be composed without actual measurements. The consumer list 
should contain the following information: process name; device 

 
Fig. 2. VSM with energy consumption and storage overview. 

Supplier Customer

Receiving twice a week              Daily shiping

Process A one shift Process B one shift Process C one shift
Storage Storage Storage Storage

c/t= 2 min 10 c/t= 4 min 15 c/t= 7 min 30
5 days C/O 2 hr  units C/O 3 hr units C/O 4 hr units

UPTIME 94% UPTIME 91% UPTIME 98%
Nom. P 30 kW Nom. P 50 kW Nom. P 2 kW

kWh/unit 2 20 kWh/unit 5 75 kWh/unit 0,3 9
kWh kWh kWh

5 days 40 minutes 105 minutes 1 day
2 minutes 4 minutes 7 minutes

II I I



label; rated power; cosᵩ; nominal current; nominal voltage and 
consumption type that can be either continuous, intermittent or 
stand by. 

Upon completion of the consumer list, traditional VSM 
should be elaborated and total nominal loads in the process 
added. As a result, elaborated VSM will show which processes 
have sufficiently high energy intensity to gain benefits from LS 
and on the other hand, to estimate the particular potential to the 
process scheduling. After completing the estimation in 
conjunction with experts from production, the VSM should be 
elaborated further. Based on the measurements, energy 
intensity or the production unit is to be found. As the energy 
usage was previously estimated, measurements can be done in 
the production process where energy usage is estimated to be 
high or VSM shows that LS can be implemented without 
investing to process capacities or intermediate stocks. 

As VSM describes product flow, the energy intensity should 
also be given per product or production unit. The unit energy 
consumption in the production process should be calculated 
based on the actual measurement. Fig. 2 shows a theoretical 
example of VSM together with data from the consumer list and 
energy intensity in the process. Energy intensity is energy 
consumption in each process to make one unit. As can be seen, 
process A C/T takes 2 minutes, i.e., one unit is completed 
during 2 minutes in the process. The total nominal power of the 
process is 30 kW and 2 kWh is consumed to make one unit. 
First intermediate stock capacity (after process A) is 10 pieces. 
It consists of 20kWh energy which is available for scheduling. 
Process B C/T is 4 minutes and the total nominal power in that 
process is 50 kW. As the second intermediate stock (after 
process B) capacity is 15 pieces, it has 75 kWh energy available 
for scheduling. Process C cycle time is 7 minutes, total nominal 
power is 2 kW and 0.3 kWh is consumed for making one unit. 
We can see that process A C/T is 2 times faster than process B 
and process C cycle time is 3.5 times faster than process A. As 
a result, we can conclude that the last process will dictate the 
whole process time and previous processes can be scheduled 
taking into account the possibilities of the last process. For that 
reason, we need to know how much time it takes to empty the 
intermediate stock before the slowest process, which we call 
buffer time (BT). In order to find BT, process C/T must be 
multiplied with an available intermediate stock capacity (ASC) 
(3) and (4). 

 
SSCMSCASC −= . (3) 

ASCTCBT ⋅= / . (4) 
 
ASC is a difference between maximum stock capacity 

(MSC) and safety stock capacity (SSC). SSC is to be defined 
by production management. 

BT shows the maximum load scheduling period in the 
process. Neglecting safety stock capacity, in our example BT is 
105 minutes. The total energy we can shift during 105 minutes 
is 95 kWh, which is the sum of stored energy in two 
intermediate stocks with 75 and 20 kWh accordingly. 

IV. EXPERIMENTAL LOAD SCHEDULING ANALYSIS 
IN A DISTRICT HEATING PLANT WITH IMPROVED 

VSM METHODOLOGY 
An example of the method above is described here. This 

example covers a district heating plant in Paide, Estonia. The 
company has one 8 MW woodchip boiler, several boilers 
fuelled with shale-oil and one CHP plant based on woodchips 
and with 8 MW thermal, 2 MW electrical output. Fig. 3 shows 
the process of the plant. Woodchip boiler stock (moving floor) 
is filled by a conveyor from the main storage (moving floor). 
The main storage is filled by the incoming tucks or a wheel 
loader. The company has 5-day storage available on site. 
Woodchip boiler stock can contain woodchips a day with boiler 
nominal load i.e. 180 m³. The conveyor between the main 
storage and the woodchip boiler stock has a maximum output 
of 50 m³/h. The conveyor from the woodchip boiler stock to the 
boiler has a strict limitation for processing output from the 
woodchip boiler. As it has no stock or daily silo available 
between the boiler and the conveyor, the output is 7.5 m³/h with 
the nominal boiler load. We start the cost reduction estimation 
from LS by modeling the process using VSM as a basis. One 
cubic meter of woodchips is taken here as one unit in VSM. 

According to the conveyor output parameters, the total output 
time is 6 days and 15.2 minutes and the value creating time is 
17.2 minutes (time when one cubic meter of woodchips is 
actually processed). Also, the VSM shows that the conveyor 
from the main to the boiler stock has excessive capacity and is 
able to process one cubic meter of woodchips approximately 
6.67 times faster than the process bottleneck, i.e. woodchip 
boiler. 

The boiler will process one cubic meter of woodchips into 
heat in 8 minutes. It can be concluded from VSM that there is a 
possibility of LS of the conveyor from the main storage to the 
boiler stock. It must be emphasized that we are dealing with a 
biomass conveyor; therefore, the following simplifications are 
used: 

a) woodchip processing by a boiler is calculated based 
on the nominal load, and boiler efficiency parameter 
is 0.8; 

b) woodchip energy intensity is approximated at 1.3 
MW/m³, in real life it can be different, based on the 
fuel type and humidity level. 

The next task is to estimate process energy intensity by 
making a consumer list behind the process. In our example, it is 
evident from VSM that the only process that can be scheduled 
is the process called “Conveyor from Main Stock to Boiler 
Stock”. Thus, we need to make a consumer list behind this 
process, which is given in Table I 

Table I shows that the total nominal power behind the 
process is 73.5 kW. The consumer list is necessary to estimate 
of the process energy demand. It is possible that some 
consumer’s nominal power is much greater than the actual 
absorbed power. Therefore it is necessary to measure the 
process energy use. 

 



 
In our example, Fluke 1735 is used for measurements of the 

process energy use and current measurements for some 
continuous consumption. There is calculated absorbed power 
with the following (5). 

 
ϕcos3 ⋅⋅⋅= ff UIP , (5) 

 
where P is absorbed power, If- phase current, Uf- phase 

voltage and cosᵩ is power factor. By using absorbed power for 
a continuous load, energy consumption was estimated as well. 
Total energy consumption for processing 1 m³ of woodchips 
was 0.96 kWh. Complete VSM for the process examined is 

given in Fig. 3. 
The VSM shows that by processing one cubic meter of 

woodchips there is consumed 0.96 kWh of electrical energy. As 
there is storage available with 180 m3 and SSC is 7.5 m3, ASC 
is 172.5 m3 using (3). By knowing process C/T behind the 
intermediate stock- it is 8 minutes per m3. There can be found 
out that BT is 23 hours by using (4). 

We can conclude that BT is sufficiently long for considering 
LS from HPP to LPP given in section II. Paide boiler plant is 
connected to a grid at low voltage line, so the grid tariffs 
applicable will be different from those given in section II. HPPP 
for the current example is 83.4 euros per MWh with 43.5 euros 
per MWh grid tariff and LPPP is 55.2 euros per MWh with 25.7 
euros per MWh grid tariff. Based on (1), we can calculate 
potential saving which is 33.8%. 

As stored energy in the process is 0.166 MWh, by 
multiplying that with HPPP, the cost for the company will be 
13.81 euros per day and saving from LS will be 4.67 euros per 
day. 

In the previous example, the VSM process is not 
complicated; however, VSM is essential in case production is 
more complicated and processes are more complex and 
dependent on each other. In that case, VSM enables cost 
savings with LS. Moreover, LS is well understandable with 
VSM for staff involved in production management and 
planning. 

 

V. CONCLUSION 
Various aspects of DSR were studied. Focus was on the price 

difference in the electricity stock market in Estonia. If the loads 
are shifted from HPP to LPP LS will enable a cost reduction of 
31%. The methodology based on VSM was introduced where 
cost savings were achieved with LS. Our analysis of 
experimental load scheduling was based on the district heating 
company in Estonia. The proposed methodology was found to 

be applicable as a straightforward tool to achieve cost savings. 
It can serve as first step for industry when implementing LS 
because of its simplicity. With minor costs, reduced time and 
lower complexity, a major cost saving can be achieved with 
simple solutions such as the method proposed. 

 
 

TABLE I 
CONSUMER LIST FOR CONVEYOR FROM MAIN STOCK TO BOILER STOCK 
Description Pn Cos ϕ In Un Consumption 
Conv. Screen to 
dist. Conv. 

15 0.8 46.9 0.4 Continuous 

Distr. conveyor 7.5 0.8 23.4 0.4 Continuous 
Floor to 
conveyor motor 

4 0.8 8.6 0.4 Continuous 

Floor to 
conveyor motor 

4 0.8 8.6 0.4 Continuous 

Hydro pack 15 0.85 28.9 0.4 Intermittent 
Hydro pack 15 0.85 28.9 0.4 Intermittent 
Screen 5.5 0.81 11.4 0.4 Continuous 
Conv. floor to 
screen 

7.5 0.8 23.4 0.4 Continuous 

 

 

 
Fig. 3 VSM for LS for Paide district heating company.. 
  

Supplier Paide Boiler Plant Customer
receiving daily continuous heat supply

Storage Storage
c/t  (min) 1.2 c/t (min) 8 c/t (min) 8

5 days C/O 0 1 day C/O 0 C/O 4 hr

Uptime 95% Uptime 95% Uptime 95%

600 (m³) Nom. P 73.5 (kW) 180 (m³)
kWh/m³ 0.96 173 (kWh) kWh/m³ 0.3

5 days 1 day
1.2 min 8 min 8 min

Conveyer from 
Storage to Boiler 

Stock
Conveyer from Boiler 

Stock to Boiler Boiler

II



REFERENCES 
[1] P. Kontogiorgos and G. P. Papavassilopoulos, "Subsidized power 

interruption for reducing peaks in energy demand: A stackelberg game," 
Communications, Control and Signal Processing (ISCCSP), 2014 6th 
International Symposium on, Athens, 2014, pp. 421-424. 
doi: 10.1109/ISCCSP.2014.6877903 

[2] G. Carpinelli, S. Khormali, F. Mottola and D. Proto, "Demand response 
and energy storage systems: An industrial application for reducing 
electricity costs. Part II: Numerical application," Power Electronics, 
Electrical Drives, Automation and Motion (SPEEDAM), 2014 
International Symposium on, Ischia, 2014, pp. 1038-1042.  
doi: 10.1109/SPEEDAM.2014.6872040 

[3] S. A. Raziei and H. Mohscnian-Had, "Optimal demand response capacity 
of automatic lighting control," Innovative Smart Grid Technologies 
(ISGT), 2013 IEEE PES, Washington, DC, 2013, pp. 1-6. 
doi: 10.1109/ISGT.2013.6497854 

[4] M. Houwing, R. R. Negenborn and B. De Schutter, "Demand Response 
With Micro-CHP Systems," in Proceedings of the IEEE, vol. 99, no. 1, 
pp. 200-213, Jan. 2011. 
doi: 10.1109/JPROC.2010.2053831 

[5] A. Rosin, I. Drovtar, S. Link, H. Hõimoja, H. Mõlder and T. Möller, 
"Analysis of large electricity consumers demand profiles to determine 
implementation of demand-side management measures”, “Tarbimise 
juhtimine ‒ suurtarbijate koormusgraafikute salvestamine ning analüüs 
juhtimise rakendamise võimaluste tuvastamiseks," Eleringi toimetised, 
pp. 1-57, 20 October 2014. 

[6] S. Mohagheghi and N. Raji, "Dynamic demand response solution for 
industrial customers," Industry Applications Society Annual Meeting, 
2013 IEEE, Lake Buena Vista, FL, 2013, pp. 1-9. 
doi: 10.1109/IAS.2013.6682583 

[7] A. Mohamed and M. T. Khan, "A review of electrical energy management 
techniques:supply and consumer side (industries)," Journal of Energy in 
Southern Africa, vol. 20, no. 3, 2009. 

[8] A. Kuhi Thalfeldt and P. Raesaar, "Possibilities and feasibility of demand-
side management in the Estonian industrial sector," in 5 th International 
Symposium "Topical Problems in the Field of Electrical and Power 
Engineering”, Kuressaare, 2008. 

[9] T. Ohno, “Toyota production system: beyond large-scale production”, 
Portland, 1988. 

[10] B. Haefnera, A. Kraemera, T. Staussa and G. Lanzaa, "Quality value 
stream mapping," in 47th CIRP Conference on Manufacturing Systemss, 
Windsor, 2014. 

[11] P. Uuemaa, J. Valtin, J. Kilter, A. Puusepp, I. Drovtar and A. Rosin, "Load 
control implementation in the energy intensive industry," Mediterranean 
Electrotechnical Conference (MELECON), 2014 17th IEEE, Beirut, 
2014, pp. 213-218. 
doi: 10.1109/MELCON.2014.6820534 

[12] I. Drovtar, P. Uuemaa, A. Rosin, J. Kilter and J. Valtin, "Using demand 
side management in energy-intensive industries for providing balancing 
power - The Estonian case study," Power and Energy Society General 
Meeting (PES), 2013 IEEE, Vancouver, BC, 2013, pp. 1-5. 
doi: 10.1109/PESMG.2013.6672418 

[13] P. Uuemaa, J. Kilter, J. Valtin, I. Drovtar, A. Rosin and A. Puusepp, 
"Cost-effective optimization of load shifting in the industry by using 
intermediate storages," Innovative Smart Grid Technologies Europe 
(ISGT EUROPE), 2013 4th IEEE/PES, Lyngby, 2013, pp. 1-5. 
doi: 10.1109/ISGTEurope.2013.6695404 

[14] AS, Nord Pool Spot, "Elspot_prices_2014_hourly_EUR," 2014. [Online]. 
Available: http://www.nordpoolspot.com/historical-market-data. 
[Accessed 20 June 2015]. 

[15] Elektrilevi OÜ, "Price list fo Elektrilevi from 01.04.2014”, “Elektrilevi 
hinnakiri alates 01.04.2014," 1th April 2014. [Online]. Available: 
www.elektrlilevi.ee/hinnakiri. [Accessed 19th June 2015] 

 


	I. INTRODUCTION
	a) conservation - energy saving is achieved through static methods;
	b) load growth - energy consumption is increased when an energy system has surplus energy production;
	c) valley filling - load is increased through the off-peak periods or keeping stable consumption;
	d) peak clipping - energy consumption is decreased in peak periods;
	e) load shifting - peak consumption is shifted from peak periods to non-peak periods;
	f) seasonal load reduction - annual energy peaks are reduced.

	II. II. ENERGY PRICE AS A DRIVER FOR LOAD SHIFTING
	III. IMPROVEMENT OF VALUE STREAM MAPPING METHODOLOGY FOR EVALUATION OF LOAD SCHEDULING POSSIBILITIES
	A. Load Scheduling at Intermediate Storage Use in Production
	B. Methodology for LS with VSM.
	a) Identify if the process is a bottleneck in production. If the process is a bottleneck, then load shifting is usually impossible without investment into the process output increase.
	b) Process cycle time (C/T) is slower or faster than its next process C/T. C/T in VSM describes how often a part or a product is completed by a process. If the process C/T is slower than the process coming next, it is possible to increase intermediate...
	c) Weather process uptime (UPTIME) is high or low, shows important information about process reliability. A low reliability process has a negative effect on LS.
	d) If stock between processes is high or low, the reasons should be found out before applying load scheduling.
	e) Change over time (C/O) is long or short. C/O is the time period for switching the process unit from one product to another. It can show also time for cold start of the process. C/O can highlight important information about the process start up time.


	IV. EXPERIMENTAL LOAD SCHEDULING ANALYSIS IN A DISTRICT HEATING PLANT WITH IMPROVED VSM METHODOLOGY
	a) woodchip processing by a boiler is calculated based on the nominal load, and boiler efficiency parameter is 0.8;
	b) woodchip energy intensity is approximated at 1.3 MW/m³, in real life it can be different, based on the fuel type and humidity level.

	V. Conclusion
	References