Training assessment: Difference between revisions

Revision as of 11:28, 31 December 2012

This page is a knowledge crystal of subtype assessment. The page identifier is Op_en5677
Moderator:Jouni (see all)
This page is a stub. You may improve it into a full page.
Upload data {{#opasnet_base_link:Op_en5677}}

This is a training assessment about an imaginary, simple case. The purpose is to illustrate assessment functionalities.

Scope

Question

What decisions are worth implementing in the training assessment?

Boundaries

Time: Year 2012 - 2020

Scenarios

Factory can reduce emissions, or continue business as usual.
School can increase health education, decrease it to save money, or continue business as usual.

Intended users

Anyone who wants to learn to make open assessments.

Participants

Main participants:
- YMAL,
- Summer workers of YMAL in 2012,
- Participants of [[Decision analysis and risk management 2013}}

Answer

Conclusions

Results

Not yet available.

Rationale

Assessment-specific data

Decisions

Decisions(-)
Obs	Decisionmaker	Decision	Option	Variable	Cell	Change	Unit	Amount
1	Factory	Cleaning.policy	Reduce emissions	exposure	Year:2020	Multiply	-	0.5
2	School	Health.promotion	Increase health education	health.impact	Year:2020	Multiply	-	0.9
3	School	Health.promotion	Promotion budget reduced	health.impact	Year:2020	Multiply	-	1.1

Probabilities

Probabilities(P)
Obs	Stakeholder	Variable	Cell	Probability
1	City of Kuopio	exposure	Cleaning.policy: Reduce emissions	0.8
2	City of Kuopio	exposure	Cleaning.policy: BAU	0.2
3	Factoy	health.impact	Health.promotion: Increase health education	0.1
4	Factoy	health.impact	Health.promotion: Promotion budget reduced	0.4
5	Factoy	health.impact	Health.promotion: BAU	0.5

Endpoints

Endpoints(-)
Obs	Stakeholder	Variable	Cell	Model	Result	Description
1	City of Kuopio	health.impact	Year:2012	Weighted sum	1000
2	City of Kuopio	health.impact	Year:2020	Weighted sum	1000
3	Citizens	health.impact	Year:2012	Weighted sum	1000
4	Citizens	health.impact	Year:2020	Weighted sum	2000	Future years are twice as important.

Variables

exposure: Training exposure
health.impact: Training health impact
training.costs: Training costs

Analyses

Decision analysis on each policy: Which option minimises the health risks?
Value of information (VOI) analysis for each policy about the major variables in the model and the total VOI.

Calculations

+ Show code - Hide code

library(OpasnetUtils)
library(xtable)
library(ggplot2)

###########################333

######################
# CheckDecisions
#####################
# Function that checks for and applies decisions for a variable if such exist. 
# Inputs:
#   * variable - ovariable to be checked
#
# For the function to actually do anything, an odecision variable that matches the name DecVariable (where "Variable" is the
# name of the variable of interest) has to be defined before running CheckDecisions. Odecision class definition can be found 
# below. The function also makes use of a preset of decision effects which match a given string constant (i.e. "Multiply"). 
#######################

CheckDecisions <- function(variable) {
	if(exists(paste("Dec", variable@name, sep = ""))) {
		
		# Initialization: Setting up a data.frame upon which to apply desired decision - option specific effect. 
		
		dec <- get(paste("Dec", variable@name, sep = "")) # Get decision variable
		dectable <- dec@dectable # Decision input table in format described on http://en.opasnet.org/w/Decision
		temp2 <- data.frame(ignoremeiamadummy = NA) # A mergeable dummy for loop initial value
		for (i in unique(as.character(dectable$Decision))) { # Decisions form new indices. Here the decision indices are merged together.
			tempdec <- dectable$Option[dectable$Decision == i]
			temp2 <- merge(temp2, tempdec)
		}
		temp2 <- temp2[!colnames(temp2) %in% "ignoremeiamadummy"] # remove dummy column
		out <- merge(variable@output, temp2) # Merge decisions with output. 
		
		# Conditions: Constructing a list of logical vectors which correspond to those rows of our new data.frame that we want to apply 
		# effects on. This will be done either by the user (as functions that take the data.frame as input and return a logical vector) 
		# while defining a custom decision or by parsing the decision table. 
		
		cond <- list()
		
		# First check if condition functions have been given. The default for the condition slot of an ovariable is a function that returns 0.
		
		if (length(dec@condition) == 1 & dec@condition[[1]](variable@output) == 0) { 
			
			# Build multiple condition vectors that correspond to a unique decision - option combination
			
			for (j in 1:nrow(dectable)) { 
				
				# In the decision table format conditions are given in the "Cell"-column separated by ";".
				
				sel1 <- strsplit(dectable[j, "Cell"], split = ";")[[1]] 
				
				# ":" defines index - location matches as a condition.
				
				sel2 <- strsplit(sel1, split = ":") # No need for lapply, since strsplit is a vectorized function and current list depth is 1.
				
				# Create a list of conditions which the decision and option specific condition vector consists of. 
				
				selection <- list() 
				for (k in 1:length(sel1)) { # For each condition separated by ";"
					if (length(sel2[[k]]) > 1) { # If ":" has been used for condition k
						locs <- strsplit(sel2[[k]][2], split = ",") # Split by "," for multiple locs per given index
						selection[[k]] <- out[, sel2[[k]][1]] %in% locs # Match our data.frame to the condition
					} #else { # Unimplemented code for (in)equality checks for numeric variables
						#if(grepl(">=")) {
							#a <- strsplit(sel1[[k]])
							#b <-  <- out[, sel2[[k]][1]]
							#selection[[k]]
						#}
					#}
				}
				
				# Match all conditions given for this decision - option combination.
				
				selection <- as.data.frame(selection)
				selection$optslice <- out[[as.character(dec[j,1])]] == dec[j,2] # We only want rows where the relevant option is in use to be affected
				selection <- as.matrix(selection)
				cond[[j]] <- apply(
					selection,
					1,
					all
				)
			}
		} else { # Otherwise use given condition functions.
			for (j in 1:length(dec@condition)) {
				cond[[j]] <- dec@condition[[j]](variable@output)
			}
		}
		
		# Effects
		
		if (length(dec@effect) == 1 & dec@effect[[1]](variable@output) == 0) {
			eff <- list()
			for (j in 1:nrow(dectable)) { 
				eff[[j]] <- EffectPreset(dectable$Change) # Returns a standard function from presets defined below.
			}
		} else {
			eff <- dec@effect
		}
		
		# Applying effects
		

		for (j in 1:nrow(dectable)) {
			out[
				cond[[j]], 
				paste(variable@name, "Result", split = "")
			] <- eff[[j]](
				out[
					cond[[j]], 
					paste(variable@name, "Result", split = "")
				], 
				dectable$Value[j]
			)
		}
		
		variable@marginal <- colnames(out) %in% c(colnames(variable@output)[variable@marginal], colnames(temp2))
		variable@output <- out
	}
	return(variable)
}

#########################

net.cost <- new("ovariable",
	name = "net.cost",
	dependencies = data.frame(
		Name = c("health.impact", "costs", "endpoints"),
		Key = c("", "", "")
	),
	formula = function(dependencies, ...) {
		ComputeDependencies(dependencies, ...)

# Using weighted sum model


		return(0)
	}
)

endpoints <- new("ovariable",
	name = "endpoints",
	data = tidy(opbase.data("Op_en5677.endpoints"), objname = "endpoints")
)

DecisionTableParser <- function(DTable){ # DTable is a data.frame
	for (i in unique(as.character(DTable$Variable))) {
		temp <- DTable[DTable$Variable == i,] # c("Decision", "Option")]
		out <- new("odecision", dectable = temp)
		assign(paste("Dec", i, sep = ""), out, envir = .GlobalEnv)
	}
}

decisions <- tidy(opbase.data("Op_en5677.decisions"), objname = "decisions")
DecisionTableParser(decisions)

Fetch2(data.frame(Name = "health.impact", Key = "iaTGnY1hgUP5tofT"))

CheckDecisions(health.impact)

probabilities <- new("ovariable",
	name = "probabilities",
	data = tidy(opbase.data("Op_en5677.probabilities"), objname = "probabilities")
)

summary(endpoints)
decisions
summary(probabilities)

References

Related files

@@ Line 47: / Line 47: @@
 <t2b name="Decisions" index="Decisionmaker,Decision,Option,Variable,Cell,Change,Observation" locations="Unit,Amount,Description" unit="-">
-Factory|Cleaning.policy|Reduce emissions|exposure|Year: 2020|Multiply|-|0.5|
+Factory|Cleaning.policy|Reduce emissions|exposure|Year:2020|Multiply|-|0.5|
-School|Health.promotion|Increase health education|health.impact|Year: 2020|Multiply|-|0.9|
+School|Health.promotion|Increase health education|health.impact|Year:2020|Multiply|-|0.9|
-School|Health.promotion|Promotion budget reduced|health.impact|Year: 2020|Multiply|-|1.1|
+School|Health.promotion|Promotion budget reduced|health.impact|Year:2020|Multiply|-|1.1|
 </t2b>

Help pages	Wiki editing • How to edit wikipages • Quick reference for wiki editing • Drawing graphs • Opasnet policies • Watching pages • Writing formulae • Word to Wiki • Wiki editing Advanced skills
Training assessment (examples of different objects)	Training assessment • Training exposure • Training health impact • Training costs • Climate change policies and health in Kuopio • Climate change policies in Kuopio
Methods and concepts	Assessment • Variable • Method • Question • Answer • Rationale • Attribute • Decision • Result • Object-oriented programming in Opasnet • Universal object • Study • Formula • OpasnetBaseUtils • Open assessment • PSSP
Terms with changed use	Scope • Definition • Result • Tool