Decision Support System Flashcards
DSS definition
information system to support decisions making
based on model simulations using various model
DSS components
- database
- model
- evaluation interface
DSS model
- inputs
- encompasses
- produces
Inputs
- natural forcing
- system state
- human forcing
- decisions
Encompasses
- hydrology
- hydraulics
- ecology
- infrastructure
Produces
- new system state to be analysed, compared to goals and constraints and inform decisions
—- maximise goals, minimise cost and satisfy constraints
system definition
- geographical area
- map water infrastructure (reservoirs, treatment plants)
- map physical and build environment
- map water demand activities
system conceptualisation
- abstraction = define influential parts of the system
- classification = unifying similar parts of system into units
- parsimony = production of the simplest model that represents all system parts
data collection
- for system state = river stage, GW, water quality
- map demand
- map infrastructure
calibration
- ID most important model outputs
- collect observations for model output
- find best param set
simulation data requirments
environmental forcing
- weather forcast for short term
- climate models long term
human forcing
- water demand from long term projections of population, urbanization, economic activity
assessing system performance
- reliability = proportion of satisfactory states
- resilience = proportion of consecutive unsatisfactory states compared all unsatisfactory states
- vulnerability = measure or how unsatisfactory unsatisfactory states are
system goals
- meet demand
- sustainability
- high standards of aquatic ecology
system constraints
- physical = reservoir size extraction limits, demand
- legal for example water framework directive
multiobjective functions aims
satisfy many, possibly contradicting objectives
pareto optimal soluton
- definition
- solution selection
Definition
- point at which non of the objective function values can be improved without degrading some of the others
- defines the pareto front
Solution selection
- turn problem to one obj func based on weighting of each func
- minimise distance from nonfeasible ideal
—- euclidean norm
—- weighted p-norm