Dear openmodders,
As I am approaching the new semester during which I will start a new class in Politecnico di Milano where I will be teaching bottom-up energy system models for planning the energy transition, there is a question that has been floating in my mind for a while.
And I thought it might be interesting to ask your opinion about that.
Is there a reference out there for clearly defining the difference between a sensitivity analysis and a scenario analysis?
Sensitivity is easier to define I guess, as its goal is clear and it is adopted in many other modelling disciplines.
But sometimes the difference with scenario analysis might become blurred, different future costs of natural gas for example.
I have an idea in my mind but I am curious to hear some of yours.
Or maybe I’m making a fool of myself and there is a precise definition that I’ve been missing all this years.
Best,
Nicolò Stevanato
Department of Energy
Politecnico di Milano
HI Nicolò I heard that the Open Energy Ontology project could not settle on a definition for "scenario" and has duly omitted that term or concept from their wider scheme. Perhaps someone involved more in the OEO could comment? Being able to specify such concepts would seem a useful precursor to answering your particular question. with best wishes, Robbie On 11.11.2023 13:34, Nicolò Stevanato wrote: > Dear openmodders, > > As I am approaching the new semester during which I will start a new > class in Politecnico di Milano where I will be teaching bottom-up > energy system models for planning the energy transition, there is a > question that has been floating in my mind for a while. > > And I thought it might be interesting to ask your opinion about that. > > Is there a reference out there for clearly defining the difference > between a sensitivity analysis and a scenario analysis? > > Sensitivity is easier to define I guess, as its goal is clear and it > is adopted in many other modelling disciplines. > > But sometimes the difference with scenario analysis might become > blurred, different future costs of natural gas for example. > > I have an idea in my mind but I am curious to hear some of yours. > > Or maybe I’m making a fool of myself and there is a precise > definition that I’ve been missing all this years. > > Best, > > Nicolò Stevanato > > Department of Energy > > Politecnico di Milano > > -- > You received this message because you are subscribed to the Google > Groups "openmod initiative" group. > To unsubscribe from this group and stop receiving emails from it, > send an email to openmod-initiat...@googlegroups.com. > To view this discussion on the web, visit > https://groups.google.com/d/msgid/openmod-initiative/43B090C5-1F69-4462-9C10-4284CCC49F4E%40hxcore.ol > [1]. > > > Links: > ------ > [1] > https://groups.google.com/d/msgid/openmod-initiative/43B090C5-1F69-4462-9C10-4284CCC49F4E%40hxcore.ol?utm_medium=email&utm_source=footer
Hi all,
Thanks for the interesting inputs. I somehow align with what Matteo and Fabian said. In fact that’s how I was planning to formulate the slide.
(I like very much Fabian’s graphical explanation, very effective)
Scenario Analysis:
Purpose:
Scenario analysis involves developing and analyzing different plausible future scenarios to assess how changes in various input parameters might affect the model's results. These scenarios are often based on different assumptions or combinations of factors that could influence the energy system.
Methodology:
Multiple scenarios are created by adjusting several input parameters simultaneously to represent different possible futures. For example, scenarios might consider variations in energy prices, technological advancements, policy changes, or demand fluctuations. The model is then run for each scenario to observe the range of potential outcomes.
Key Features:
Scenario analysis helps decision-makers understand the range of possible outcomes under different conditions and can be valuable for strategic planning and risk management.
Sensitivity Analysis:
Purpose:
Sensitivity analysis, focuses on understanding how changes in specific input parameters impact the model's results. It helps identify which factors have the most significant influence on the model's outputs and which are less critical.
Methodology:
Sensitivity analysis involves varying one input parameter at a time while keeping others constant to observe the resulting changes in the model outputs. The goal is to identify parameters that, when altered, have a substantial impact on the results.
Key Features:
Sensitivity analysis provides insights into the relative importance of different parameters and helps prioritize efforts to gather more accurate data for those that have a greater impact on the model's outcomes.
Nicolò
From: Fabian Neumann
Sent: 13 November 2023 08:58
To: openmod initiative
Subject: Re: [openmod-initiative] Scenario vs Sensitivity Analysis
Hi all,
My mental model to distinguish the two is relatively similar to Matteo's.
- scenario analysis is narrative-based, usually there is just a handful of scenarios (e.g. high wind investment cost / low solar investment costs and vice versa)
- sensitivity analysis is about the response of the model outputs to a change in one (local) or multiple (global) inputs
An excellent overview I read was Section 2.2. in Will Usher's PhD thesis: https://discovery.ucl.ac.uk/id/eprint/1504608/1/Usher-W__e-thesis_200715_V2.0.0_final.pdf
"
Global sensitivity analysis is a family of techniques used to determ-
ine the influence of a model input upon a model output. Sensitivity
analysis is often conflated with Uncertainty Analysis (UA) and un-
certainty propagation (UP), but these are both distinct steps within
a full meta-study of a model’s response to uncertainty. An UA is the
process of quantifying the uncertainties in a model’s inputs through
expert elicitation, statistical analyses and so on. Then, the aim of UP
is to quantify to what extent uncertainty exists in the outputs of a
model by using a technique, such as Monte Carlo sampling (or al-
ternative), to propagate uncertainty through a model. However, the
aim of sensitivity analysis is to, independently of the uncertainty of a
parameter, determine each parameter’s influence upon the model out-
put. So to paraphrase Hamby, (1994), an uncertainty analysis ranks
parameters according to their importance, and a sensitivity analysis
ranks parameters according to sensitivity
"
To view this discussion on the web, visit https://groups.google.com/d/msgid/openmod-initiative/1c7288e9-0c2e-4229-b50f-cc22fa8d5c6cn%40googlegroups.com.
Emission scenarioA plausible representation of the future development of emissionsof substances that are radiatively active (e.g., greenhouse gases(GHGs) or aerosols) based on a coherent and internally consistentset of assumptions about driving forces (such as demographicand socio-economic development, technological change, energyand land use) and their key relationships. Concentration scenarios,derived from emission scenarios, are often used as input to a climatemodel to compute climate projections.
To view this discussion on the web, visit https://groups.google.com/d/msgid/openmod-initiative/1c7288e9-0c2e-4229-b50f-cc22fa8d5c6cn%40googlegroups.com.
Hi all,
In my perception, the discussion about how we define the term scenario is kind of the default discussion at the beginning of each collaborative project with people from different domains. For this reason, I once also tried to create a categorization and added “sensitivity analyses” as synonym for a certain kind of “scenarios”. Maybe sharing this with you is helpful, maybe not 😊
Best,
Kiên
Best,
Kiên
To view this discussion on the web, visit https://groups.google.com/d/msgid/openmod-initiative/CAKNUXC3OP_8DREM0O6Gca5JnFQ2XT4vzEpVcaXrE6HWDxbU06Q%40mail.gmail.com.
To view this discussion on the web, visit https://groups.google.com/d/msgid/openmod-initiative/a16b80de-66a6-0ae6-c480-37fa9589f9b9%40boku.ac.at.
Hello openmoders,
Does anyone has the power/rights to block the flow of emails from Kathaleen Prabel which have nothing to do with energy system modelling?
Best,