03-02-2025 © Deep Kiran, IIT Roorkee (2025) 1

EEN/L-671:
RESTRUCTURED POWER
SYSTEMS
LECTURE 8: Organization of Electricity
Markets
03-02-2025 © Deep Kiran, IIT Roorkee (2025) 2

What is the difference between average

cost and marginal cost?
• Both costs are expressed as ₹ / unit produced
• Marginal cost (MC) reflects cost of last unit produced
• Average cost (AC) reflects cost of all units already
produced
• Fixed costs (FC) do not reflect in MC
03-02-2025 © Deep Kiran, IIT Roorkee (2025) 3

Relation between average cost and

marginal cost?
MC AC

• For low production levels, MC is

smaller than AC because of effect
of FC
• For higher production levels, MC
is higher than AC
• MC curve intersects AC curve at
its minimum

y
03-02-2025 © Deep Kiran, IIT Roorkee (2025) 4

Costs: Numerical example

Q Fixed cost Variable cost Total cost AFC AVC AC MC
0 1000 0 1000
10 1000 120 1120 100 12 112 22
20 1000 440 1440 50 22 72 42
30 1000 960 1960 33.33333 32 65.33333 62
40 1000 1680 2680 25 42 67 82
50 1000 2600 3600 20 52 72 102
60 1000 3720 4720 16.66667 62 78.66667 122
70 1000 5040 6040 14.28571 72 86.28571 142
80 1000 6560 7560 12.5 82 94.5 162
90 1000 8280 9280 11.11111 92 103.1111 182
100 1000 10200 11200 10 102 112 202
110 1000 12320 13320 9.090909 112 121.0909 222
120 1000 14640 15640 8.333333 122 130.3333 242
130 1000 17160 18160 7.692308 132 139.6923 262
140 1000 19880 20880 7.142857 142 149.1429 282
150 1000 22800 23800 6.666667 152 158.6667 302
160 1000 25920 26920 6.25 162 168.25 322
170 1000 29240 30240 5.882353 172 177.8824 342
180 1000 32760 33760 5.555556 182 187.5556 362
190 1000 36480 37480 5.263158 192 197.2632 382
200 1000 40400 41400 5 202 207 402
 03-02-2025 © Deep Kiran, IIT Roorkee (2025) 5

Costs: Numerical example

45000

40000

Fixed cost Variable cost Total cost

35000

30000

25000
Cost

20000

15000

10000

5000

0
0 10 20 30 40 50 60 70 80 90 100 110 120 130 140 150 160 170 180 190 200
MW
 03-02-2025 © Deep Kiran, IIT Roorkee (2025) 6

Costs: Numerical example

450

400

350
AFC AVC AC MC

300

250
Cost

200

150

100

50

0
0 10 20 30 40 50 60 70 80 90 100 110 120 130 140 150 160 170 180 190 200
MW
03-02-2025 © Deep Kiran, IIT Roorkee (2025) 7

Short-run and long-run

• Many variables of production
• All factors do not affect the output in the same way
• Some factors affect in immediate way (short-run) while
others take long time to be effective (long-run)
• In short-run, some of the production factors are fixed
• Long-run provides time sufficient enough to adjust various
factors
• Short-run: output depends on single factor of production,
rest are fixed
• Long-run: All factors of production can be adjusted
03-02-2025 © Deep Kiran, IIT Roorkee (2025) 8

Short-run and long-run

• Long run cost calculation is complex
• More degrees of freedom available
• Analogy: Increasing power transfer capability of a
transmission corridor:
• Short-run: install a capacitor bank
• Long-run: build a parallel line
• Decisions about power system operation are short-run
• Decisions about power system planning are long-run
03-02-2025 © Deep Kiran, IIT Roorkee (2025) 9

Example
• Demonstrate that the marginal production cost is equal to
the average production cost for the value of the output
that minimizes the average production cost.
03-02-2025 © Deep Kiran, IIT Roorkee (2025) 10

Example
• A firm’s short-run cost function for the production of
gizmos is given by the following expression: 𝐶(𝑦) =
10𝑦 2 + 200𝑦 + 100000
• Calculate the range of output over which it would be profitable for
this firm to produce gizmos if it can sell each gizmo for ₹2400.
Calculate the value of the output that maximizes this profit.
• Repeat these calculations and explain your results for the
case in which the short-run cost function is given by
𝐶(𝑦) = 10𝑦 2 + 200𝑦 + 200000
03-02-2025 © Deep Kiran, IIT Roorkee (2025) 11

Relation between long-run and short-run

average costs
AC in short-run Long-run AC
AC in short-run big factory
small factory
Price per vehicle

`
AC in short-run
medium factory

Vehicle manufactured per day

03-02-2025 © Deep Kiran, IIT Roorkee (2025) 12

Long-run and short-run

• The long-run cost curve provides a lower envelope for all
short run average cost curves.
• This peculiar relationship exists because of flexibility
associated while obtaining long-run cost curve
• As the firm moves from small size factory to medium size factory,
the cost per vehicle starts reducing till a certain point
• As long as the average production cost decreases, the
product is said to display economies of scale.
03-02-2025 © Deep Kiran, IIT Roorkee (2025) 13

Perfectly competitive market

• Atomicity: Large number of small players
• Complete information: consumers and sellers know the
prices set by all firms
• Free entry: No firm has barrier for entry into or exit out of
market

All producers are price takers in perfectly competitive

market
03-02-2025 © Deep Kiran, IIT Roorkee (2025) 14

Firm’s supply decision under perfect

competition
max[𝜋 × 𝑦 − 𝑐 𝑦 ]
03-02-2025 © Deep Kiran, IIT Roorkee (2025) 15

Revenues of the firm

• Total revenue:
• πxq
• Average revenue:
• (π x q)/q
• Marginal revenue:
• change in revenue from sale of addition output
• Marginal revenue is ‘π’ in perfectly competitive case
 03-02-2025 © Deep Kiran, IIT Roorkee (2025) 16

Firm’s supply decision under perfect

competition • For q1, marginal revenue is
greater than marginal cost.
• Hence tendency to increase output
Cost and revenue

• For q2, reduced production

MC leads to saving in marginal
cost which is greater than
marginal revenue.
• Hence, tendency to reduce the
output

MC2
AC

π*=MR1=MR2
π*=AR=MR
AVC

MC1

q1 q* q2 Quantity
03-02-2025 © Deep Kiran, IIT Roorkee (2025) 17

Imperfect Competition
• The firms decide upon the quantity produced so as to
have direct effect on market price
• Prices can be manipulated by:
• Withholding the quantity
• Raising the asking price
03-02-2025 © Deep Kiran, IIT Roorkee (2025) 18

Common forms of imperfect competition

• Monopoly: only one seller
• Oligopoly: finite small number of sellers
• Monopsony: only one buyer
• Oligopsony: small number of buyers of good
03-02-2025 © Deep Kiran, IIT Roorkee (2025) 19

Monopoly
• Monopoly firm is a sole seller of product
• Fundamental cause of monopoly is barrier to entry:
• Key resource for production is owned by a single firm
• Government gives exclusive rights to a single firm
• Costs of production make a single producer more efficient than
large number of producers
• An industry is a natural monopoly when a single firm can
supply a good to an entire market at a smaller cost than
two or more firms
03-02-2025 © Deep Kiran, IIT Roorkee (2025) 20

Competitive firm Vs. monopoly

Competitive Firm Monopoly

• Small size relative to the • Single producing firm

market • Ability to influence the
• Price taker price of its output
• Faces a horizontal • Faces a market demand
demand curve curve (downward sloping)
 03-02-2025 © Deep Kiran, IIT Roorkee (2025) 21

Competitive firm Vs. monopoly

𝜋
𝜋

Demand curve
Demand curve
𝜋 ∗= 𝑀𝑅
Marginal revenue curve

Marginal revenue curve

q q

Competitive firm Monopoly

03-02-2025 © Deep Kiran, IIT Roorkee (2025) 22

Profit maximization by monopolist

price

MC

B
π*

AC
A
Demand

MR

q1 q* q2 Quantity
03-02-2025 © Deep Kiran, IIT Roorkee (2025) 23

Oligopoly
• Modeling of oligopoly:
1. Cournot Model: firms decide quantity that they produce
2. Stackelberg Model: firms decide quantity that they
produce
3. Bertrand Model: firms decide the price

Perfect
Monopoly Oligopoly
Competition
03-02-2025 © Deep Kiran, IIT Roorkee (2025) 24

Attendance
• MS Teams: l6ahq8m
• Please ensure 75% of attendance for ETE.

• Please finalize a paper from IEEE TEMPR in a team of two.

Please get it confirmed from me at the earliest!
• Brief introduction to the work done in the chosen paper.
• What was the situation before the authors did the work? What was the
motivation for the authors that led them to do this work?
• Describe the methodology of the work in the chosen paper. Explain the
optimization model.
• Describe the key numeric results given in the paper which justify the
findings of the paper.
• What do you think about the actual implementation/ validation/ testing
of the work proposed in real life world/ test bed?

• Friday: 10 – 11 am
03-02-2025 © Deep Kiran, IIT Roorkee (2025) 25

Cournot model
𝑌 = 𝑦1 + 𝑦2 + ⋯ + 𝑦𝑛

max[𝑦𝑖 𝜋 𝑌 − 𝑐 𝑦𝑖 ]
𝑦𝑖

𝑑
𝑦𝑖 𝜋 𝑌 − 𝑐 𝑦𝑖 =0
𝑑𝑦𝑖

𝑑𝜋 𝑌 𝑑𝑐 𝑦𝑖
𝜋 𝑌 + 𝑦𝑖 =
𝑑𝑦𝑖 𝑑𝑦𝑖
03-02-2025 © Deep Kiran, IIT Roorkee (2025) 26

Cournot model
𝑦𝑖 𝑌 𝑑𝜋 𝑌 𝑑𝑐 𝑦𝑖
𝜋 𝑌 1+ =
𝑌 𝑑𝑦𝑖 𝜋 𝑌 𝑑𝑦𝑖

𝑦𝑖
⇒ 𝑠𝑖 =
𝑌

𝑠𝑖 𝑑𝑐 𝑦𝑖
⇒𝜋 𝑌 1− =
𝜀 𝑌 𝑑𝑦𝑖

𝑠𝑖
: should be high for existence of market power
𝜀 𝑌
03-02-2025 © Deep Kiran, IIT Roorkee (2025) 27

Readings from Cournot model

• If market share of a firm is not negligible, it maximizes
profit by setting its production at a level where its marginal
cost is less than the market price.
• The equation suggests that a low price elasticity of
demand and a degree of market concentration facilitate
the exercise of market power.
03-02-2025 © Deep Kiran, IIT Roorkee (2025) 28

Optimization
• Science of making best decisions
• Used in an extraordinary range of applications.
• Production, Manufacturing, Scheduling
• Distribution, supply chains
• Human resource planning (e.g., crew scheduling)
• Healthcare
• Designing technologies (e.g., networks, semiconductors, airplanes)
• Process design
• Airlines and trucking
• Retailing
• Oil and gas
• Electric power
• Financial services
• Agriculture
• Telecommunications
• Internet
• MUCH MORE.
03-02-2025 © Deep Kiran, IIT Roorkee (2025) 29

Optimization Preliminaries
• In any optimization problem
• An objective function to be optimized
• Some constraints to be satisfied: equality and inequality constraints
• Objectives:
• Cost, loss, voltage stability margin, etc.
• Constraints:
• Power balance – equality constraint
• Voltage, generation, line flows – inequality constraints
03-02-2025 © Deep Kiran, IIT Roorkee (2025) 30

Optimization techniques
• Two broad types:
• Simplex method (solves only linear programming problems)
• Gradient based methods
• First and second order gradient methods (Newton’s methods)
• Interior point methods
• Evolutionary algorithms
• Genetic algorithm
• Particle swarm optimization
• Differential evolution
03-02-2025 © Deep Kiran, IIT Roorkee (2025) 31

Linear programming (LP)

• Reasons for simplicity and high speed of LP.
• Graphical representation of a two variable problem.
max 3𝑥1 + 5𝑥2
3𝑥1 + 2𝑥2 ≤ 18
0 ≤ 𝑥1 ≤ 4
0 ≤ 𝑥2 ≤ 6
03-02-2025 © Deep Kiran, IIT Roorkee (2025) 32

Simple example
• A farmer has recently acquired a 110 hectares piece of land.
He has decided to grow wheat and millet on that land. Due to
the quality of the sun and the region’s excellent climate, the
entire production of Wheat and Millet can be sold. He wants to
know how to plant each variety in the 110 hectares, given the
costs, net profits and labour requirements according to the data
shown below:
Variety Cost (₹/Hec) Net Profit (₹/Hec) Man-days/Hec
Wheat 100 50 10
Millet 200 120 30

• The farmer has a budget of ₹10,000 and availability of 1,200

man-days during the planning horizon. Find the optimal
solution and the optimal value.
03-02-2025 © Deep Kiran, IIT Roorkee (2025) 33

Simple example
• Chateau Maxim produces three different types of wines,
Red, Rose and White, using three different types of
grapes planted in its own vineyard. The amount of each
grape necessary to produce a unit amount of each wine,
the daily production of each grape, and the profit of selling
a unit of each wine is given below. How much of each
wines should one produce to maximize the profit? We
assume that all wines produced can be sold.
Wines (ton/unit)
Grapes Red White Rose Supply (ton/day)
Pinot Noir 2 0 0 4
Gamay 1 0 2 8
Chardonnay 0 3 1 6
profit (K₹/unit) 3 4 2
03-02-2025 © Deep Kiran, IIT Roorkee (2025) 34

Organization of Electricity Markets:

Contents
03-02-2025 © Deep Kiran, IIT Roorkee (2025) 35

Modes of energy trading

• Bilateral contracts
• Customized long term contracts (one to one)
• Trading over the counter (OTC) (one to one)
• Electronic exchange (anonymous)
• Pool Trading
• ‘Equilibrium’ obtained in a systematic way
 03-02-2025 © Deep Kiran, IIT Roorkee (2025) 36

Timeline of Contracts
Seller
Seller
Long term Day Ahead
bilateral Market (run by
Buyer transactions SO or PX)
Pool exchange
Buyer
or Pool
Dispatch

Seller
Seller

Short term Real Time

bilateral Market (run by
transactions SO)
Real Time
Electronic Buyer
exchange or Buyer Market
trader
03-02-2025 © Deep Kiran, IIT Roorkee (2025) 37

Why BT don’t take place near real time?

• Gaps between generation and load should be filled up
very quickly
• Market mechanisms are too slow
• Role of system operator is a must
03-02-2025 © Deep Kiran, IIT Roorkee (2025) 38

Why do people get into BT ahead of real

time?
• Price near real time is volatile
• Most participants want more certainty
• Reduce risk by trading ahead of the spot market
• Forward markets and derivative markets help reduce risks