Dynamics of supply-chain and market volatility of networks

1
Dynamics of supply-chain and market volatility of
networks
WP5
Fernanda Strozzi Cattaneo University-LIUC Italy
2
Work package number 5 Start date or starting event Start date or starting event Start date or starting event Start date or starting event Month 0 Month 0 Month 0
Work package title Dynamics of supply-chain and market volatility of networks Dynamics of supply-chain and market volatility of networks Dynamics of supply-chain and market volatility of networks Dynamics of supply-chain and market volatility of networks Dynamics of supply-chain and market volatility of networks Dynamics of supply-chain and market volatility of networks Dynamics of supply-chain and market volatility of networks Dynamics of supply-chain and market volatility of networks
Participant ID QMUL JRC COLB MASA LIUC LIUC NESA GEME
Person-months per participant 6 3 18 4 24 24 Ad hoc Ad hoc
Objectives To understand and measure how the volatility, in the time series of energy market spot prices affects congestion and its links to frequencies and length of blackouts trends in European synchronously connected grids. To connect the Electricity grid with supply chain networks and electricity market spot prices. To develop and implement an algorithm for Early Warning Detection of blackouts.
3
WP5 Tasks overview
Coupling models Task5.5
EWDS of Blackouts T5.4
Electricity price Model T5.1
Interaction Risk T5.6
Electric power Model T5.1
Supply chain Model T5.1, T5.5
Correlations(T5.2) Analysis(T5.3)
Energy spot prices Volatility
Blackouts Volatility
4
Deliverables

• D5.1. Report on supply-chain logical model by
  means of the Petri nets formalism (M12),
  completed.
• D5.2. Report on market dynamics model (M12),
  completed.
• D5.3. Report (paper) on Cross Recurrence
  Quantification Analysis between markets
  volatility and the dynamics of power systems
  dynamic (M24), 50 Done
• D5.4.Report (paper) on coupled market dynamics -
  power systems- supply chains (M30)
• D5.5. Report on early warning detection
  algorithm and suggestions on how implement it in
  real systems (M36)
• EWDS developed for a one level supply-chain.

5
D5.1 (M12) Impact of the electric power supply on
a logistic-production system
Fault generation Model Monte Carlo
One supply chain level Petri Net
service level
System Dynamic Model
6
D5.1Supply-Chain and electric model

• Faults generation model
• Model of Medium Voltage (2400-34500 volts) Power
  Distribution System in Presence of faults and
  restoration events.
• Monte Carlo Simulation to generate a Random Walk
  to simulate the faults.
• Simulation of the protection device intervention
  and reconfiguration of the system.
• System Dynamic logical model
• Identification of the important variable in one
  supply chain level and their relationships.
• Model of one supply chain level
• Discrete model using Petri Net

7
D5.2 (M12)Electricity Markets and Spot Price
Models

• The available studies are classified in terms of
  the applied methodology . The proposed models can
  be broadly divided into three classes
• Statistical technical analysis, simple
  autoregressive models.
• Econometric more sophisticate models with
  jumps, peak over the threshold
• and regime switching. Other models are
  focused on price
• volatility evaluation.
• Structural fundamental methods, including the
  impacts of important
• physical and economic factors on the spot
  price (Economic Cycles)
• The available models are mostly either for
  univariate or uniequational analysis.
• Not agreement on the models to utilize and on
  the main variables to be considered

8
D5.2 (M12)Electricity Markets and Spot Price
Models

• Dynamic Factor Models
• Stock and Watson (2002a,b, 1999)
• Factor Model (Principal Component) to identify
  latent (not measured)variables Dynamical model
  (Dynamical Factor) to study the relationships
  between variables.
• They can cope with many variables without running
  into scarce degrees of freedom.
• They can manage large data set at a high
  disaggregate level.
• They have not been explored for the electricity
  price dynamics

9
D5.2 (M12) Future Work
Electricity price Factor Dynamic Model
Fault generation Model Monte Carlo
One supply chain level Petri Net
?
Global service activity
service level
System Dynamic Model
10
D5.3 (M24) RQA analysis of electricity prices
and blackout
Recurrence Plots (RPs) represent the distance
between state space points of a time series. RPs
use state space reconstruction techniques
(normally delayed vectors of only one measured
variable) RQA extracts quantitative information
from Recurrence Plots, in terms of several
parameters DET(determinism), RR( recurrence),
LAM (laminarity)
x
t
11
D5.3 (M24)
RQA parameters are able to distinguish between
spot electricity prices dynamics and Gaussian
linear correlated noise with the same
autocorrelation function (the same FFT)
RQA parameters gives a new measure of the
dispersion of data (volatility) and dynamical
information.
12
D5.3 (M24)RQA analysis of electricity prices and
blackout
1 Application of non-linear time series
analysis techniques to the Nordic spot
electricity market data. F. Strozzi, E.Gutiérrez,
C. Noè, T. Rossi, M.Serati and J.M. Zaldívar.
LIUC Paper 200, october 2007. 2 Application
of RQA to Financial Time Series, F. Strozzi, J.M.
Zaldivar, J. Zbilut, Second International
workshop on Recurrence Plot, Siena, 10-12
September 2007. 3 Measuring volatility in the
Nordic spot electricity market using Recurrence
Quantification Analysis. F. Strozzi, E.Gutiérrez,
C. Noè, T. Rossi, M.Serati and J.M. Zaldívar .
Submitted to Physica D
13
D5.5 (M36)Early Warning Detection System (EWDS)
for Blackouts
Divergence control

• Divergence gives a filtered measure of the
  acceleration of the measured variable.
• Divergence can be obtained analytically from the
  model of the system (the trace of its Jacobian).
• Divergence can be reconstructed on-line using
  only one measured variable

On-line Safety and optimization of chemical
reactions controlling temperature to prevent
runaway reactions On-line Trading startegy
applied to high frequencies stock
exchange better results than RSI (Relative
Strenght Index)
14
D5.5 (M36) Early Warning Detection Algorithm for
Blackouts
Bullwip control in supply-chain application
off-line and comparison with a proportional
control
D0
O1
D1O0
Bullwhipi var(Di)/var(Oi)
2 Strozzi, F., Zaldivar, J.M., Noè, C., 2007,
The Control of Local Stability and of the
bullwhip effect in a supply chain. International
Journal of Production Economics (In press). 3
Caloiero, G., Strozzi, F., Zaldívar, J.M., 2007.
A supply chain as a serie of filter or amplifiers
of the bullwhip effect. International Journal of
Production Economics (Accepted).
15
D5.5 (M36) Early Warning Detection Algorithm for
Blackouts
Bullwip control in supply-chain on-line
application and comparison with a proportional
control
Cost reduction with the new control technique
based on the divergence of the system in case of
a periodic noisy demand
4 Strozzi, F., Noè, C., and Zaldivar, J.M.
2007, Control and on-line optimization of a
supply chain. (In preparation).
16
LIUC Collaborations

• Qeen Mary (Physica A)
• JRC (Physica A, Physica D)
• COLB (under discussion)
• MASA (under discussion)

LIUC Gender Action

• 2 female PhD students started to work on
• Models of Supply Chain
• Ranking Risk in Supply Chain