|
|
Farminder Anand
PhD Candidate
311
Ferst Drive Nw.
Atlanta, GA 30332
School of Chemical & Biomolecular Engineering
Georgia Institute of Technology
Email:
farminder.anand[@]chbe.gatech.edu
Phone: 404-385-2149
|
EDUCATION & AWARDS
 M.S. (Non-thesis), Chemical Engineering (2008)
Georgia Institute of
Technology
-
PSE Fellowship, 2004-2009, Georgia Tech
M.S. (Non-thesis), Quantitative Computational Finance (2008)
Georgia Institute of
Technology
B.Tech., Pulp and Paper
Engineering (2004)
Indian Institute of Technology,
IIT. Roorkee, India
-
Student Scholarship (2001-2004)
-
Institute Silver
medal, for academic excellence and best project of the year (2004)
RESEARCH INTERESTS
Broadly speaking my research interest is in the following areas:
o
Process design modeling and evaluation of various Biorefinery configurations
o Design of Experiment Strategies (Bayesian/Classical)
o
Bayesian Statistics
o
Decision Analysis, Risk Management
o Multi Stage decision making under uncertainty
The following two subsections give further insight about my research:
Bayesian
Decision Framework (BDF):
We are trying to
put together various tools a R&D investment decision maker can use of and
integrating them together so as to formulate an integrated ‘Bayesian Decision
Framework’. These tools though common in various other fields, such as
‘Decision Analysis’, ‘Financial Engineering’, ‘Decision Support Systems’,
etc., are not very widely used for R&D investment decision making in
‘Chemical Engineering’. A sneak peek on the framework is shown in figure 1.
The application of these tools in their respective fields is well studied but
they have not seen dawn in ‘Chemical Engineering’ and hence their application
in our realm presents both great challenges and opportunities.
Figure 1 Bayesian Decision Making Framework
Biorefinery
Options:
[2
Conference Paper, 1 Technical Report]
“Biorefinery” is
a relatively new term. Composed of the terms ‘Bio’ and ‘Refinery’, it refers
to the conversion of biomass feedstock into a host of valuable chemicals and
energy with minimal waste and emissions.
Concerns over
the reliance of the U.S. economy on crude oil, the reliance on foreign
sources for the majority of this commodity, and the recent high volatility of
crude oil prices have led to an increased interest in diversifying the U.S.
energy base. The higher crude oil prices make processes that utilize
renewable energy sources such as biomass, wind and solar power competitive in
the marketplace. Biorefineries, integrated production systems for materials
and energy from agricultural or silvicultural inputs, are considered capable
of meeting part of the need for renewable resources. Pulp and paper mills
have existing infrastructure to receive, store, and handle woody biomass
residuals, which can be used to produce fuels, and thus they have an initial
capital base for a biorefinery and a potential leverage for increased value
from these inputs. However, there are many different configurations for such
biorefineries, and it is important to analyze the alternatives systematically
to uncover their different strengths and weaknesses before major research and
development and capital expenditures are made.
In order to
uncover the tradeoffs of the various configurations, we did a carbon balance
for the few chosen cases [.pdf].
Figure 2 briefly describes the type of questions we intended to answer in
this case study.
Figure 2 Biorefinery Options in a Pulp/Paper
mill.
The various
issues which we had to face during this simple though not seemingly
challenging case study, and the results hence obtained played a key
motivating role for our work on ‘Bayesian Decision Framework’. The specific
configuration, which we intend to use to test the practicality of the BDF is
shown in Figure 3.
Figure 3 Process flow diagram of integrated
Biorefinery with both Bio-chemical & Thermo-chemical options.
Decision
Oriented Bayesian Design of Experiments:
[1
Conference Paper, 1 Journal Paper(under preparation)]
Experimental design is a fundamental problem in
science and engineering. Traditional ‘Design of Experiment’ (DOE) approaches
focus on minimization of variance. In this work, we propose two new
“decision-oriented” DOE approaches, which takes into account how the generated
data, and subsequently the developed model, will be used in decision making.
By doing so, the variance will be distributed in a manner such that its
impact on the targeted decision making will be minimal. Our results show that
the new decision-oriented experiment design approach significantly
outperforms the standard D-optimal design approach. The new design method
should be a valuable tool when experiments are conducted for the purpose of
making R&D decisions.
Expert
Opinion Elicitation to obtain Prior distributions:
[1
Conference Paper, 1 Journal Paper(under preparation)]
Expert Opinion elicitation is a special technique
used in the fields of ‘Decision Analysis’ and ‘Risk Analysis’. We have
incorporated and demonstrated their use in the evaluation of developing
processes. Global Sensitivity Analysis (GSA) is special method used
rigorously in environmental engineering for understanding the importance key
uncertain factors in the complex models. As the evaluation of biorefinery
processes involves the similar level of complex models and high uncertainty,
their application is kind of intuitive though we have never found them to be
used along with the process design modeling approaches in Chemical
Engineering. Using Expert Opinions provides the decision maker with the prior
distributions on the model parameters, which can both be used for quantifying
the uncertainties in the process design models and as priors for further
designing Experimental strategies.
CONFERENCE PAPERS
Farminder S. Anand, Matthew Realff and Jay
H. Lee, Decision Oriented Bayesian
Design of Experiments (ADCHEM, July 12-15, 2009, Koç University, Istanbul,
Turkey)
Farminder S.
Anand, Matthew Realff and Kristiina Iisa, Quantifying
uncertainties in Borefinery options- Using Expert Opinions, 2008 AiChE
Annual Meeting, Philadelphia, PA, Nov 16- Nov 21, 2008.
Farminder S. Anand, Matthew
Realff, Kristiina Iisa and Charles E. Courchene, Evaluating Biorefinery Options in a Pulp and Paper
Mill: Carbon Balances, TAPPI, 2006 Engineering, Pulping &
Environmental Conference, Atlanta, GA, Nov. 5- Nov. 8,2006. [.pdf]
Kristiina Iisa, Charles
Courchene, Farminder Anand, Matthew Realff, Jim Frederick, Process Systems Considerations in
Forest Biorefineries with Thermochemical Processing of Wood Wastes,
AIChE 2005 Annual Conference, Cincinnati, OH, Oct. 30-Nov.4,2005. [.pdf]
TECHNICAL
REPORT
Larson, E. D.,
S. Consonni, R. E. Katofsky, K. Iisa, W. J. Frederick, C. Courchene, F.
Anand, and M. Realff (2006): "A Cost-Benefit Assessment of
Gasification-Based Biorefining in the Kraft Pulp and Paper Industry,Volume 4,
Preliminary Biorefinery Analysis with Low- Temperature Black Liquor
Gasification."
JOURNAL PUBLICATIONS
Farminder S. Anand, Jay H. Lee and Matthew
Realff, Optimal Decision oriented
Bayesian design of experiments (Under preparation)
Farminder S. Anand, Matthew Realff and Jay
H. Lee, Using Expert Opinions and
Copula’s to improve R&D decisions
(Under preparation)
|