Package ‘utility’

August 29, 2013
Type Package Title Construct, Evaluate and Plot Value and Utility Functions Version 1.1 Date 2013-03-03 Author Peter Reichert <peter.reichert@eawag.ch> with contributions by Nele Schuwirth <nele.schuwirth@eawag.ch> Maintainer Peter Reichert <peter.reichert@eawag.ch> Description Construct and plot objective hierarchies and associated value and utility functions. Evaluate the values and utilities and visualize the results as colored objective hierarchies or tables. Visualize uncertainty by plotting median and quantile intervals within the nodes of objective hierarchies. Get numerical results of the evaluations in standard R data types for further processing. License GPL (>= 2) NeedsCompilation no Repository CRAN Date/Publication 2013-03-03 08:41:25

R topics documented:
utility-package . . . . . . . . . evaluate . . . . . . . . . . . . . evaluate.utility.aggregation . . . evaluate.utility.conversion.intpol evaluate.utility.conversion.parfun evaluate.utility.endnode.cond . . evaluate.utility.endnode.discrete evaluate.utility.endnode.intpol1d evaluate.utility.endnode.intpol2d . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 1 . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 3 10 12 13 15 16 18 19 21

2 evaluate.utility.endnode.parfun1d . plot.utility.aggregation . . . . . . plot.utility.conversion.intpol . . . plot.utility.conversion.parfun . . . plot.utility.endnode.cond . . . . . plot.utility.endnode.discrete . . . . plot.utility.endnode.intpol1d . . . plot.utility.endnode.intpol2d . . . plot.utility.endnode.parfun1d . . . print.utility.aggregation . . . . . . print.utility.conversion.intpol . . . print.utility.conversion.parfun . . . print.utility.endnode.cond . . . . . print.utility.endnode.discrete . . . print.utility.endnode.intpol1d . . . print.utility.endnode.intpol2d . . . print.utility.endnode.parfun1d . . summary.utility.aggregation . . . . summary.utility.conversion.intpol . summary.utility.conversion.parfun summary.utility.endnode.cond . . summary.utility.endnode.discrete . summary.utility.endnode.intpol1d summary.utility.endnode.intpol2d summary.utility.endnode.parfun1d updatepar . . . . . . . . . . . . . updatepar.utility.aggregation . . . updatepar.utility.conversion.intpol updatepar.utility.conversion.parfun updatepar.utility.endnode.cond . . updatepar.utility.endnode.discrete updatepar.utility.endnode.intpol1d updatepar.utility.endnode.intpol2d updatepar.utility.endnode.parfun1d utility.aggregate.add . . . . . . . . utility.aggregate.cobbdouglas . . . utility.aggregate.max . . . . . . . utility.aggregate.min . . . . . . . utility.aggregate.mix . . . . . . . utility.aggregate.mult . . . . . . . utility.aggregation.create . . . . . utility.calc.colors . . . . . . . . . utility.conversion.intpol.create . . utility.conversion.parfun.create . . utility.endnode.cond.create . . . . utility.endnode.discrete.create . . utility.endnode.intpol1d.create . . utility.endnode.intpol2d.create . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .

R topics documented: . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 22 24 27 30 33 34 36 37 39 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60 61 63 64 65 66 67 68 70 71 72 74 75 82 83 85 87 89 91 94

utility-package

3

utility.endnode.parfun1d.create . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 96 utility.fun.exp . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 98 utility.get.colors . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 100 Index 101

utility-package

Construct, Evaluate and Plot Value and Utility Functions

Description Construct and plot objective hierarchies and associated value and utility functions. Evaluate the values and utilities and visualize the results as colored objective hierarchies or tables. Visualize uncertainty by plotting median and quantile intervals within the nodes of the objective hierarchy. Get numerical results of the evaluations in standard R data types for further processing. Details Package: Type: Version: Date: License: utility Package 1.1 2013-03-03 GPL (>= 2)

An objective hierarchy and an associated value or utility function is constructed by constructing the nodes of the hierarchy starting from the end nodes and proceeding to the higher hierarchies. Five types of end nodes are distinguished: End nodes of the class utility.endnode.discrete define a value or utility function for an attribute that has a finite number of discrete numeric or non-numeric levels. End nodes of the classes utility.endnode.intpol1d and utility.endnode.parfun1d implement single-attribute value or utility functions that accept a continuous argument. The first of these functions allows the user to specify attribute-value pairs and performs linear interpolation between these points. The second function allows the user to specify any parameteric function that is implemented as a function in R. End nodes of the class utility.endnode.intpol2d implement interpolated value or utility functions that are based on two attributes. Finally, end nodes of the class utility.endnode.cond implement value or utility functions that assign different value or utility functions to a finite set of attribute combinations. These end nodes can be implemented by using the following constructors. utility.endnode.discrete.create utility.endnode.intpol1d.create utility.endnode.parfun1d.create utility.endnode.intpol2d.create utility.endnode.cond.create To advance to higher hierarchical levels, values or utilities at lower levels must be aggregated to the next higher level. This is done ab aggregation nodes of the class utility.aggregation. Such

4 nodes can be implemented by using the following constructor: utility.aggregation.create

utility-package

Finally, to provide decision support under uncertainty, values at an adequate level of the objectives hierarchy must be converted to utilities by accounting for the risk attitude of the decision maker. Similar to the single-attribute value or utility functions, this can either be done by linear interpolation with a node of the class utility.conversion.intpol or by using a parametric funciton in a node of the class utility.conversion.parfun. These conversion nodes can be implemented by the constructors: utility.conversion.intpol.create utility.conversion.parfun.create The definition of the objective hierarchy and the associated value and utility function can then be listed or visualized by using the generic functions print summary plot which automaticall call the implementation corresponding to the node specified as the first argument: print.utility.endnode.discrete print.utility.endnode.intpol1d print.utility.endnode.parfun1d print.utility.endnode.intpol2d print.utility.endnode.cond print.utility.aggregation print.utility.conversion.intpol print.utility.conversion.parfun summary.utility.endnode.discrete summary.utility.endnode.intpol1d summary.utility.endnode.parfun1d summary.utility.endnode.intpol2d summary.utility.endnode.cond summary.utility.aggregation summary.utility.conversion.intpol summary.utility.conversion.parfun plot.utility.endnode.discrete plot.utility.endnode.intpol1d plot.utility.endnode.parfun1d plot.utility.endnode.intpol2d plot.utility.endnode.cond plot.utility.aggregation

utility-package plot.utility.conversion.intpol plot.utility.conversion.parfun The value or utility function can then be evaluated by applying the generic function evaluate that again calls automatically the corresponding class-specific function evaluate.utility.endnode.discrete evaluate.utility.endnode.intpol1d evaluate.utility.endnode.parfun1d evaluate.utility.endnode.intpol2d evaluate.utility.endnode.cond evaluate.utility.aggregation evaluate.utility.conversion.intpol evaluate.utility.conversion.parfun

5

This function requires the provision of observed or predicted attributes of the valued system and returns the corresponding values or utilities of all nodes of the hierarchy. These results can then be visualized by providing them to the generic function plot in addition to the definition of the objective hierarchy stored in the variable corresponding to the highest node of the hierarchy. Again, this function automatically calls the correct class-specific implementation (the root of the hierarchy will be an aggregation or a conversion node, not an end node): plot.utility.aggregation plot.utility.conversion.intpol plot.utility.conversion.parfun This proceedure guarantees easy handling with the simple commands print, summary, evaluate, and plot and the specific function descriptions provided above are only required to check advanced attributes. Author(s) Peter Reichert <peter.reichert@eawag.ch> with contributions by Nele Schuwirth <nele.schuwirth@eawag.ch> Maintainer: Peter Reichert <peter.reichert@eawag.ch> References Short description of the package: Reichert, P., Schuwirth, N. and Langhans, S., Constructing, evaluating and visualizing value and utility functions for decision support, Environmental Modelling & Software, in press.

6 Textbooks on the use of utility and value functions in decision analysis:

utility-package

Keeney, R. L. and Raiffa, H. Decisions with Multiple Objectives. Cambridge University Press, 1993. First published in 1976. Eisenfuehr, F., Weber, M. and Langer, T., Rational Decision Making, Springer, Berlin, 2010. Examples
# define discrete end node for width variability # (attribute "widthvariability_class" with levels "high", # "moderate" and "none") widthvar <utility.endnode.discrete.create( name.node = "width variability", attrib.levels = data.frame(widthvariability_class= c("high","moderate","none")), u = c(1, .4125, ), names.u = c("u.high","u_moderate","u.none"), required = FALSE, utility = FALSE) # define 1d interpolation end node for bed modification with # riprap # (attribute "bedmodfract_percent" with levels from to 1 ) bedmod_riprap <utility.endnode.intpol1d.create( name.node = "bed modification riprap", name.attrib = "bedmodfract_percent", range = c( ,1 ), x = c( ,1 ,3 ,1 ), u = c(1, .775, .5625, .24), required = FALSE, utility = FALSE) # define 1d interpolation end node for bed modification with # other material # (attribute "bedmodfract_percent" with levels from to 1 ) bedmod_other <utility.endnode.intpol1d.create( name.node = "bed modification other", name.attrib = "bedmodfract_percent", range = c( ,1 ), x = c( ,1 ,3 ,1 ), u = c(1, .775, .5625, ), required = FALSE, utility = FALSE) # define combination end node for bed modification

"other")). .create( name.endnode.1 ).levels = data. nodes = list(bedmod_riprap. utility = FALSE) # define 1d interpolation end node for bank modification with # permeable material # (attribute "bankmodfract_percent" with levels from to 1 ) bankmod_perm <utility.1 ).bankmod_imperm).3 .6 .attrib = "bankmodfract_percent".775. .3 .cond.attrib = "bankmodfract_percent". required = FALSE. .1 . utility = FALSE) # define 2d interpolation end node for riparian zone width # (attributes "riparianzonewidth_m" and "riparianzonewidth_m") 7 .8667.675. .frame(bedmodtype_class= c("riprap".levels = data.node = "bed modification".1 ). required = FALSE. attrib. u = c(1. range = c( .node = "bank modification perm". utility = FALSE) # define combination end node for bank modification # (attributes "bankmodtype_class" and "bankmodfract_percent") bankmod <utility.utility-package # (attributes "bedmodtype_class" and "bedmodfract_percent") bedmod <utility.6 . .frame(bankmodtype_class= c("perm".create( name.create( name.cond. required = FALSE. .bedmod_other).24."imperm")).4125.1 ).5625.node = "bank modification imperm".endnode.create( name. attrib. name. .endnode. name. u = c(1.24). x = c( . ).intpol1d.1 . range = c( .node = "bank modification". nodes = list(bankmod_perm. x = c( . required = FALSE. utility = FALSE) # define 1d interpolation end node for bank modification with # impermeable material # (attribute "bankmodfract_percent" with levels from to 1 ) bankmod_imperm <utility.endnode.intpol1d.

". required = FALSE.node = "ecomorphology". name.levels = data.1).25.16).y=c( .5. .aggregation. ."artificial")). required = FALSE) # define aggregation node for ecomorphological state morphol <utility."w_min".create( name. lead = 1. names. ."w_cobbdouglas").endnode. .y=c(15. nodes = list(widthvar. nodes = list(riparzone_width. required = TRUE) # print individual definitions print(widthvar) print(bedmod) .aggregation.8 utility-package riparzone_width <utility.riparzone). list(x=c( . .fun = "utility.bankmod. list(x=c( . .mix".bedmod.1).3 )).1.fun = "utility."riparianzonewidth_m"). . name.attrib = c("riverbedwidth_m". # "seminatural" and "artificial") riparzone_veg <utility.intpol2d.25.25. par = c( . ).25.frame(riparianzoneveg_class= c("natural".y=c(5.y=c(3 .node = "riparian zone".16).create( name.1. ).15)). . utility = FALSE) # define discrete end node for riparian zone vegetation # (attriute "riparianzoneveg_class" with levels "natural".node = "riparian zone width".cobbdouglas".15)). u = c(1.16)."w_bankmod".aggregate.riparzone_veg).create( name.endnode. .16).16)."w_bedmod".aggregate.6. name. list(x=c( .15. isolines = list(list(x=c( . attrib. utility = FALSE) # define aggregation node for riparian zone riparzone <utility.3 ))).c( .node = "riparian zone veg.5625.discrete.2."w_riparzone".create( name.1 ."seminatural". par = c(1. )).par = c("w_widthvar". u = c( . ranges = list(c( . "w_add".

5 . 5 . par=c(w_add=1. "seminatural") "high".res_rehab) rownames(res_both) <.attrib=attrib_channelized.frame(widthvariability_class bedmodtype_class bedmodfract_percent bankmodtype_class bankmodfract_percent riverbedwidth_m riparianzonewidth_m riparianzoneveg_class = = = = = = = = = = = = = = = = "none". "imperm".w_min= .utility-package 9 # print all definitions print(morphol) # plot objectives hierarchy with attributes plot(morphol) # plot individual nodes: plot(widthvar) plot(widthvar.data.c("channelized"."rehabilitated") plot(morphol.2)) plot(bedmod_other) plot(bankmod) #plot(riparzone_width) # plot selected node definitions of a hierarchy plot(morphol.attrib=attrib_rehab) res_both <. 5.w_cobbdouglas= )) res_rehab <. 15.par=c(u_moderate= .u=res_channelized) plot(morphol. "riprap". 15. 7 .evaluate(morphol.evaluate(morphol.nodes=c("width variability". "bank modification")) # evaluate value function for data sets and plot colored hierarchies # and table attrib_channelized <.data. "bed modification other".attrib=attrib_channelized) res_channelized_add <.frame(widthvariability_class bedmodtype_class bedmodfract_percent bankmodtype_class bankmodfract_percent riverbedwidth_m riparianzonewidth_m riparianzoneveg_class attrib_rehab <. 2 .type="nodes". "natural") res_channelized <. 1 . "imperm".u=res_channelized_add) plot(morphol.u=res_rehab) . "riprap".rbind(res_channelized.evaluate(morphol.

2 evaluate attrib_channelized_unc <.sampsize).mean=1 . bankmodtype_class = rep("imperm".sampsize).uref=res_channelized_unc) evaluate Evaluate Node and Associated Hierarchy Description Generic function to calculate values or utilities at all nodes of a hierarchy for given levels of the attributes. bedmodfract_percent = rnorm(sampsize.sampsize).sd=5).u=res_rehab_unc.frame( widthvariability_class = c("moderate".attrib=attrib_channelized_unc) res_rehab_unc <. bankmodfract_percent = rnorm(sampsize.u=res_channelized_unc) plot(morphol.sampsize).type="table") # consideration of uncertain attribute levels (higher uncertainty for # predicted state after rehabilitation than for observed channelized state): sampsize <. bankmodtype_class = rep("imperm".mean=7 .u=res_rehab_unc) plot(morphol.1.sd=5)..uref=res_channelized) plot(morphol.mean=2 . .sd=15).data.5)+1]) res_channelized_unc <.sampsize).sd=2). .mean=1 . .mean=5 . bedmodfract_percent = rnorm(sampsize."high")[rbinom(sampsize.1.sampsize).evaluate(morphol.attrib=attrib_rehab_unc) plot(morphol. riparianzonewidth_m = rep(5.10 plot(morphol. bedmodtype_class = rep("riprap".5)+1]. bedmodtype_class = rep("riprap"."seminatural")[rbinom(sampsize.) .frame( widthvariability_class = rep("high"."artificial")[rbinom(sampsize.data. riverbedwidth_m = rnorm(sampsize.mean=5 .u=res_both.sd=2). riverbedwidth_m = rep(1 .sampsize).evaluate(morphol. riparianzoneveg_class = c("seminatural". Usage evaluate(x. .1. bankmodfract_percent = rnorm(sampsize. riparianzoneveg_class = c("natural".sd=5). riparianzonewidth_m = rnorm(sampsize.5)+1]) attrib_rehab_unc <..u=res_rehab.

endnode....reichert@eawag.endnode.conversion.create.intpol1d. utility.create. Springer. . and Langhans. 1993. utility. Weber. 2010. and Langer. First published in 1976. P.endnode. F. Examples # see help(utility) # for examples. M..evaluate Arguments x .create. R. Rational Decision Making. Cambridge University Press. in press.create. evaluating and visualizing value and utility functions for decision support. Environmental Modelling & Software. Author(s) Peter Reichert <peter. Berlin.aggregation.intpol2d. S. T. Constructing. N.create.create to create the nodes to be evaluated. Schuwirth. 11 attribute levels have to be provided as an additional argument attrib. parameter values can optionally be provided as an additional argument par. node to be evaluated. utility.conversion..ch> References Short description of the package: Reichert.endnode. utility.endnode. utility. See Also See utility.parfun1d. Decisions with Multiple Objectives.intpol. Eisenfuehr. L. utility.discrete. and Raiffa..cond. Textbooks on the use of utility and value functions in decision analysis: Keeney.parfun.create.create. utility. Value Data frame with results of values or utilities at all nodes of the hierarchy for all provided sets of attribute levels. H.

(optional) labelled numeric parameter vector providing parameters to modify the value or utility function before evaluation. node to be evaluated.aggregation Evaluate Node and Associated Hierarchy Description Calculate values or utilities at all nodes of a hierarchy for given levels of the attributes. N..ch> References Short description of the package: Reichert. Environmental Modelling & Software.aggregation evaluate. R. 1993. Constructing.aggregation’ evaluate(x.12 evaluate. Rational Decision Making.. P. Cambridge University Press. M. L. F. Berlin. S. and Raiffa. currently no other arguments are implemented or passed further.) Arguments x attrib par .. Author(s) Peter Reichert <peter..utility. . par = NA. 2010. Schuwirth.utility. First published in 1976. Textbooks on the use of utility and value functions in decision analysis: Keeney.. and Langer.. Weber. in press.. .. evaluating and visualizing value and utility functions for decision support. numeric vector with labelled components providing the levels of a single set of attributes or data frame for which each row provides such a set of attributes.reichert@eawag. and Langhans. Usage ## S3 method for class ’utility. Eisenfuehr. Decisions with Multiple Objectives. attrib. Value Data frame with results of values or utilities at all nodes of the hierarchy for all provided sets of attribute levels. H. Springer. T.

utility.parfun1d.endnode.conversion.intpol’ evaluate(x.cond.endnode. numeric vector with labelled components providing the levels of a single set of attributes or data frame for which each row provides such a set of attributes.utility.utility. utility..endnode.endnode.conversion. utility. .intpol.conversion.create to create other nodes.conversion.parfun. print. par = NA. utility. utility. attrib.intpol See Also utility.aggregation or summary. and plot. currently no other arguments are implemented or passed further. .evaluate..aggregation to plot the node and utility.discrete.intpol2d.aggregation to print its definition.utility. node to be evaluated.utility.utility. 13 evaluate. Usage ## S3 method for class ’utility.create.) Arguments x attrib par .create to create the node.create.conversion.intpol1d.create.create. (optional) labelled numeric parameter vector providing parameters to modify the value or utility function before evaluation.create.aggregation. utility.endnode...create.intpol Evaluate Node and Associated Hierarchy Description Calculate values or utilities at all nodes of a hierarchy for given levels of the attributes. Examples # see help(utility) # for examples.

Eisenfuehr. Textbooks on the use of utility and value functions in decision analysis: Keeney. Cambridge University Press.intpol2d. F. utility. S. 2010.intpol or summary.parfun1d.intpol1d.conversion.create to create other nodes. See Also utility.conversion.endnode.intpol. N.intpol Data frame with results of values or utilities at all nodes of the hierarchy for all provided sets of attribute levels.cond.create. utility.reichert@eawag.endnode. Springer.utility. Environmental Modelling & Software.aggregation.ch> References Short description of the package: Reichert.conversion.14 Value evaluate.intpol to plot the node and utility.. Rational Decision Making. Examples # see help(utility) # for examples. .endnode. L.conversion. Constructing. R.conversion. Decisions with Multiple Objectives.create. First published in 1976.endnode. and Langhans.utility.create. in press.utility. print.utility. Schuwirth. and Raiffa. H. utility. utility.create to create the node.conversion. M. and plot. T... 1993. and Langer. utility.create.intpol to print its definition. evaluating and visualizing value and utility functions for decision support.create. Berlin. utility. Weber.parfun. Author(s) Peter Reichert <peter..endnode. P.create.discrete.

currently no other arguments are implemented or passed further... Rational Decision Making.parfun’ evaluate(x... R. P. Usage ## S3 method for class ’utility.evaluate.. . Environmental Modelling & Software. and Langer.conversion. and Raiffa. Textbooks on the use of utility and value functions in decision analysis: Keeney.. Eisenfuehr.) Arguments x attrib par . First published in 1976.conversion. L. . S. T. evaluating and visualizing value and utility functions for decision support.utility.. Constructing. H. node to be evaluated. 2010. M. Author(s) Peter Reichert <peter. and Langhans. F. Decisions with Multiple Objectives. Berlin. Weber. par = NA. attrib.reichert@eawag.utility.conversion.parfun Evaluate Node and Associated Hierarchy Description Calculate values or utilities at all nodes of a hierarchy for given levels of the attributes. Cambridge University Press. 1993. in press. Springer.. (optional) labelled numeric parameter vector providing parameters to modify the value or utility function before evaluation. N. numeric vector with labelled components providing the levels of a single set of attributes or data frame for which each row provides such a set of attributes.ch> References Reichert. Value Data frame with results of values or utilities at all nodes of the hierarchy for all provided sets of attribute levels. Schuwirth.parfun 15 evaluate.

utility.aggregation. par = NA.utility. utility. utility.utility.endnode.endnode.endnode. Usage ## S3 method for class ’utility.) Arguments x attrib par .utility. (optional) labelled numeric parameter vector providing parameters to modify the value or utility function before evaluation.create. node to be evaluated. currently no other arguments are implemented or passed further. evaluate.cond.discrete.cond’ evaluate(x. print. utility.aggregation.create.16 See Also evaluate.aggregation or summary.endnode.create.create.cond utility..endnode. numeric vector with labelled components providing the levels of a single set of attributes or data frame for which each row provides such a set of attributes.. utility.create.parfun1d.utility. and plot.create.cond Evaluate Node and Associated Hierarchy Description Calculate values or utilities at the node for given levels of the attributes.endnode. utility.aggregation to print its definition.create to create other nodes. attrib.aggregation to plot the node and utility.endnode.intpol1d.intpol.conversion.utility.intpol2d. Examples # see help(utility) # for examples.. . .create to create the node..endnode.

and plot. and Langhans. P. L. Berlin. in press.aggregation or summary. H.create. F..discrete.utility. Constructing. .cond Value 17 Numeric vector of results of values or utilities at the node for all provided sets of attribute levels. Schuwirth. utility. utility.conversion. Decisions with Multiple Objectives. N. See Also utility. and Raiffa.create to create other nodes. Springer.endnode. 1993. utility. S. 2010.evaluate..reichert@eawag.intpol2d. and Langer. Cambridge University Press.utility. utility.parfun1d.intpol.utility.conversion.create.create to create the node.create..create. evaluating and visualizing value and utility functions for decision support. Eisenfuehr.aggregation to plot the node and utility.endnode. Rational Decision Making.aggregation to print its definition.aggregation.create. R.utility.endnode. First published in 1976. utility. T. Author(s) Peter Reichert <peter. Weber.aggregation. M. utility. print..create. Examples # see help(utility) # for examples.ch> References Short description of the package: Reichert. Textbooks on the use of utility and value functions in decision analysis: Keeney.endnode.intpol1d.endnode.parfun. Environmental Modelling & Software.

T. 1993. F.. evaluating and visualizing value and utility functions for decision support..endnode. currently no other arguments are implemented or passed further. and Raiffa. and Langhans.ch> References Short description of the package: Reichert. 2010. Textbooks on the use of utility and value functions in decision analysis: Keeney. Author(s) Peter Reichert <peter. First published in 1976... N. Value Numeric vector of results of values or utilities at the node for all provided sets of attribute levels. S.utility.endnode. (optional) labelled numeric parameter vector providing parameters to modify the value or utility function before evaluation. numeric vector with labelled components providing the levels of a single set of attributes or data frame for which each row provides such a set of attributes. L. Berlin.reichert@eawag.discrete Evaluate Node Description Calculate values or utilities at the node for given levels of the attributes. M..discrete’ evaluate(x.. Eisenfuehr. node to be evaluated.utility. par = NA. . in press. Usage ## S3 method for class ’utility. and Langer. Environmental Modelling & Software. Schuwirth. Weber. H.discrete evaluate.. R. Constructing. P. Cambridge University Press. . Springer. Rational Decision Making..18 evaluate. attrib.) Arguments x attrib par . Decisions with Multiple Objectives.endnode.

endnode.intpol1d Evaluate Node Description Calculate values or utilities at the node for given levels of the attributes.utility.intpol1d’ evaluate(x.cond. print.aggregation to plot the node and utility. 19 evaluate.create to create other nodes.) Arguments x attrib par . .create to create the node.aggregation or summary.create.intpol1d.utility..conversion.endnode.utility.aggregation. utility. (optional) labelled numeric parameter vector providing parameters to modify the value or utility function before evaluation. .create. par = NA..evaluate. currently no other arguments are implemented or passed further.endnode.intpol2d.endnode. Usage ## S3 method for class ’utility. Examples # see help(utility) # for examples. utility.create.aggregation.parfun. attrib. numeric vector with labelled components providing the levels of a single set of attributes or data frame for which each row provides such a set of attributes.intpol1d See Also utility.create.intpol.endnode.utility. node to be evaluated. utility. utility..conversion.create. utility.endnode. utility.utility. and plot.aggregation to print its definition..endnode.create.parfun1d.

. P.ch> References Short description of the package: Reichert.endnode.. Author(s) Peter Reichert <peter. S. and Langer. Examples # see help(utility) # for examples. and plot.parfun.aggregation to print its definition.. M.create to create the node.create. Schuwirth.utility.parfun1d. utility.endnode.aggregation.create.endnode.utility. utility. in press.utility. Berlin.aggregation to plot the node and utility.create. utility.reichert@eawag..utility. evaluating and visualizing value and utility functions for decision support. print.aggregation.aggregation or summary.create.20 Value evaluate.intpol2d. N.intpol1d Numeric vector of results of values or utilities at the node for all provided sets of attribute levels. utility.create. Springer.endnode. L.conversion. utility. R.cond.. Cambridge University Press. Weber. 1993. Rational Decision Making. Decisions with Multiple Objectives. See Also utility. Environmental Modelling & Software. Eisenfuehr.create to create other nodes.conversion.endnode. 2010. and Raiffa. Constructing. T.create. and Langhans. utility. First published in 1976. Textbooks on the use of utility and value functions in decision analysis: Keeney. H. F.discrete.intpol.

M. node to be evaluated. par = NA.endnode. Weber. 1993. (optional) labelled numeric parameter vector providing parameters to modify the value or utility function before evaluation..endnode..intpol2d’ evaluate(x. and Raiffa. Springer. evaluating and visualizing value and utility functions for decision support. ..intpol2d 21 evaluate. Author(s) Peter Reichert <peter. R. currently no other arguments are implemented or passed further. numeric vector with labelled components providing the levels of a single set of attributes or data frame for which each row provides such a set of attributes. L. 2010. Eisenfuehr. Value Numeric vector of results of values or utilities at the node for all provided sets of attribute levels.. Decisions with Multiple Objectives.reichert@eawag.utility.) Arguments x attrib par ..utility.intpol2d Evaluate Node Description Calculate values or utilities at the node for given levels of the attributes. Schuwirth. F.. attrib. H.evaluate. P. . T. S. Usage ## S3 method for class ’utility. First published in 1976. Berlin. and Langhans. Environmental Modelling & Software.. Cambridge University Press. Constructing. Textbooks on the use of utility and value functions in decision analysis: Keeney.. in press. Rational Decision Making. N.ch> References Short description of the package: Reichert. and Langer.endnode.

print.aggregation to print its definition.) Arguments x attrib par .22 See Also evaluate. currently no other arguments are implemented or passed further.endnode. attrib. utility.parfun1d’ evaluate(x. (optional) labelled numeric parameter vector providing parameters to modify the value or utility function before evaluation.endnode.create.create. utility.utility.utility..discrete.endnode.intpol.create to create other nodes..create.cond. utility.create.intpol1d. evaluate..aggregation or summary.create to create the node.parfun.utility. utility. utility.conversion. numeric vector with labelled components providing the levels of a single set of attributes or data frame for which each row provides such a set of attributes.parfun1d utility.aggregation. Examples # see help(utility) # for examples.aggregation.conversion. Usage ## S3 method for class ’utility.utility.endnode..endnode.parfun1d.endnode.endnode.aggregation to plot the node and utility. par = NA.create. . node to be evaluated.create. utility. . and plot.parfun1d Evaluate Node Description Calculate values or utilities at the node for given levels of the attributes.utility.

create.aggregation to print its definition.endnode. See Also utility.aggregation. evaluating and visualizing value and utility functions for decision support.utility. Decisions with Multiple Objectives.cond.create to create other nodes.utility. utility.endnode.conversion. and Langer. utility.evaluate. . M. First published in 1976. P. Cambridge University Press.intpol1d. and plot. and Raiffa. T. Constructing. utility. Environmental Modelling & Software.. F. 1993. R. utility. utility.aggregation. Weber. print.endnode. utility. Rational Decision Making.create to create the node.endnode. Author(s) Peter Reichert <peter. S. Examples # see help(utility) # for examples. H.utility. and Langhans.intpol. Schuwirth.parfun1d Value 23 Numeric vector of results of values or utilities at the node for all provided sets of attribute levels.intpol2d..aggregation or summary. L.create. Eisenfuehr.. 2010..utility.create. Springer.reichert@eawag.aggregation to plot the node and utility. Berlin.discrete. N.create.conversion.ch> References Short description of the package: Reichert. in press.parfun.create.endnode.create. Textbooks on the use of utility and value functions in decision analysis: Keeney.

(optional) vector or data frame with elements or columns labelled according to the nodes of the hierarchy containing values or utilities. this will be the complete output or an output row of the function evaluate.2. this will be the complete output or an output row of the function class ’utility.nodes cex. (optional) vector or data frame with elements or columns labelled according to the nodes of the hierarchy containing values or utilities. NA.8). It is then used to color-code the boxes of the hierarchy representing value nodes or the table.6. u uref par type nodes col gridlines main cex. . Usage ## S3 method for plot(x. This input is only considered if the argument type is specified to be either "hierarchy" or "table".main cex.utility.attrib f. 1. "table".aggregation. Typically. 1. "".2. c( . NA.4.calc.utility.aggregation plot.utility.24 plot. .attrib . .aggregation Plot Node Definition or Underlying Objectives Hierarchy Description Plot node definition or underlying objective hierarchy. . . NA. then the median and quantile boxes are plotted for value nodes or the expected utility for utility nodes unless the argument main contains as many elements as the number of rows of u..aggregation’ = = = = = = = = = = = = = = NA. TRUE. "node". separate hierarchies with color-coded boxes for value nodes are produced for all rows of u. Typically. c("hierarchy".. uref . "nodes"). In the latter case. If u is a data frame with more than one row and the argument type is equal to "hierarchy".reaches f.colors(). 1.2.nodes with. utility.) Arguments x u node to be plotted.

the medians and colored boxes indicating 90% credibility or occurrence ranges are plotted at all nodes. (optional) numeric vector of levels at which gridlines are plotted in node definitions.reaches f. "node" or "nodes". This argument only affects the output if the argument type was indicated to be either "table" or "nodes". Otherwise.plot. (optional) specifies the type of plot to be produced. It is then used to color-code the upper part of the boxes of the hierarchy to allow for a comparison with the results provided by the argument u which are shown in the lower part of the boxes. "table": produces a table with color-coded results for values or utilities if these values are provided by the argument u. type nodes col gridlines main cex.nodes cex. This attribute is only used if the argument type is specified to be either "node" or "nodes". "nodes": produces plots of node definitions for all nodes defined by the attribute nodes. "hierarchy": produces a plot of the objectives hierarchy including color-coded results for values or utilities if these values are provided by the arguments u and/or uref. (optional) character vector of colors to be used to color the interval between zero and unity in equidistant sections (use repetitions of the same color if you want to have a non-equidistant color-coding). (optional) character vector specifying the nodes for which the definitions will be plotted or which will be considered in a table.aggregation before passing the results of this function to this plotting routine.utility.utility.main cex. (optional) title(s) of the plot. "node": produces a plot of the definition of the current node. (optional) fraction of the width of the plot reserved for the row labels of the table if the argument type is equal to "table". To color-code hierarchies or tables for different parameter values.utility. Note that this affects only the node definitions plotted if the argument type is specified to be "node" or "nodes". This attribute is only used for value nodes and if values are provided by the arguments u and/or uref. This input is only considered if the argument type is specified to be "hierarchy". Options: "hierarchy".attrib f. (optional) scaling factor for title of the plot.nodes . (optional) fraction of the height of the plot reserved for the column labels of the table if the argument type is equal to "table". "table". the parameters have to be passed to evaluate. (optional) scaling factor for node labels used in the plot. (optional) scaling factor for attribute labels used in the plot.aggregation. If the argument type is equal to "hierarchy" and the a vector of titles with the same length as the number of rows of the argument u is provided. The default value of NA indicates that all nodes will be plotted.aggregation 25 evaluate. a color-coded hierarchy is plotted for each row of u. par (optional) labelled numeric parameter vector providing parameters to modify the value or utility function before plotting the node.

Examples # see help(utility) # for examples. evaluating and visualizing value and utility functions for decision support. Textbooks on the use of utility and value functions in decision analysis: Keeney. 1993.. L.utility.aggregation for how to evaluate the node. S. and Langhans. See Also See utility. Note that the plotting routines plot.colors for an example of how to construct color schemes and utility.aggregation (optional) indicates if attributes should be listed if the argument type is equal to "hierarchy". Berlin. T.colors for how to get colors for specifed value levels. M.26 with.. and Langer. Author(s) Peter Reichert <peter.utility.reichert@eawag. Springer.calc.aggregation. Constructing. Rational Decision Making. Note plot. Environmental Modelling & Software.attrib .get. First published in 1976. R. additional arguments passed to the R plotting routine. Cambridge University Press.parfun are exactly the same so that all hierarchies can be plotted with exactly the same commands irrespective of the type of the top-level node.. F. and Raiffa. in press.conversion. Schuwirth. 2010.ch> References Short description of the package: Reichert. H.. See utility. Eisenfuehr. .. N.utility.intpol plot.conversion.create for how to construct such a node and evaluate.utility. P.. Decisions with Multiple Objectives. Weber.

"node".nodes with.nodes cex.plot. If u is a data frame with more than one row and the argument type is equal to "hierarchy".attrib . utility.intpol’ = = = = = = = = = = = = = = NA. (optional) vector or data frame with elements or columns labelled according to the nodes of the hierarchy containing values or utilities.conversion. . uref . 1.utility.conversion. . .8). "table". c("hierarchy".intpol Plot Node Definition or Underlying Objectives Hierarchy Description Plot node definition or underlying objective hierarchy. TRUE. Typically.calc.utility. "nodes").intpol 27 plot.2.2.) Arguments x u node to be plotted.4. "". Usage ## S3 method for plot(x. (optional) vector or data frame with elements or columns labelled according to the nodes of the hierarchy containing values or utilities.attrib f. Typically.main cex. NA..conversion. this will be the complete output or an output row of the function evaluate. NA. .. 1. this will be the complete output or an output row of the function class ’utility. 1.aggregation. It is then used to color-code the boxes of the hierarchy representing value nodes or the table.utility.6.2. separate hierarchies with color-coded boxes for value nodes are produced for all rows of u. c( . NA.colors().reaches f. then the median and quantile boxes are plotted for value nodes or the expected utility for utility nodes unless the argument main contains as many elements as the number of rows of u. . u uref par type nodes col gridlines main cex. This input is only considered if the argument type is specified to be either "hierarchy" or "table". In the latter case.

Otherwise. (optional) scaling factor for attribute labels used in the plot. (optional) character vector specifying the nodes for which the definitions will be plotted or which will be considered in a table. "node": produces a plot of the definition of the current node. type nodes col gridlines main cex. "table". This attribute is only used for value nodes and if values are provided by the arguments u and/or uref. "nodes": produces plots of node definitions for all nodes defined by the attribute nodes. (optional) specifies the type of plot to be produced.utility.28 plot. the medians and colored boxes indicating 90% credibility or occurrence ranges are plotted at all nodes. "hierarchy": produces a plot of the objectives hierarchy including color-coded results for values or utilities if these values are provided by the arguments u and/or uref.attrib f.conversion.intpol evaluate. To color-code hierarchies or tables for different parameter values. (optional) title(s) of the plot. The default value of NA indicates that all nodes will be plotted. (optional) fraction of the height of the plot reserved for the column labels of the table if the argument type is equal to "table". "node" or "nodes". This attribute is only used if the argument type is specified to be either "node" or "nodes". a color-coded hierarchy is plotted for each row of u. If the argument type is equal to "hierarchy" and the a vector of titles with the same length as the number of rows of the argument u is provided. Note that this affects only the node definitions plotted if the argument type is specified to be "node" or "nodes". This input is only considered if the argument type is specified to be "hierarchy".utility. "table": produces a table with color-coded results for values or utilities if these values are provided by the argument u. (optional) character vector of colors to be used to color the interval between zero and unity in equidistant sections (use repetitions of the same color if you want to have a non-equidistant color-coding). (optional) fraction of the width of the plot reserved for the row labels of the table if the argument type is equal to "table".aggregation before passing the results of this function to this plotting routine.nodes .aggregation.main cex. (optional) numeric vector of levels at which gridlines are plotted in node definitions. This argument only affects the output if the argument type was indicated to be either "table" or "nodes".nodes cex. Options: "hierarchy".utility. (optional) scaling factor for title of the plot. the parameters have to be passed to evaluate. (optional) scaling factor for node labels used in the plot. par (optional) labelled numeric parameter vector providing parameters to modify the value or utility function before plotting the node. It is then used to color-code the upper part of the boxes of the hierarchy to allow for a comparison with the results provided by the argument u which are shown in the lower part of the boxes.reaches f.

calc.utility. Berlin. and Langer.intpol for how to evaluate the node. L. Constructing. and Raiffa..utility. Author(s) Peter Reichert <peter.reichert@eawag. 2010.intpol with.conversion.colors for an example of how to construct color schemes and utility. Schuwirth. R.utility. Examples # see help(utility) # for examples. T. Springer. S.plot. Weber. additional arguments passed to the R plotting routine.create for how to construct such a node and evaluate.conversion. First published in 1976.intpol.. P.get. M..colors for how to get colors for specifed value levels.parfun plot.utility. and Langhans. See utility. See Also See utility. F.aggregation are exactly the same so that all hierarchies can be plotted with exactly the same commands irrespective of the type of the top-level node.ch> References Short description of the package: Reichert. . H. evaluating and visualizing value and utility functions for decision support. 1993. in press.. Decisions with Multiple Objectives. Environmental Modelling & Software. Textbooks on the use of utility and value functions in decision analysis: Keeney.attrib . Cambridge University Press.. Eisenfuehr. Note 29 (optional) indicates if attributes should be listed if the argument type is equal to "hierarchy".conversion. N. Note that the plotting routines plot.. Rational Decision Making.conversion.

utility.nodes cex. Usage ## S3 method for plot(x. . Typically. .6.utility..attrib f.2. . separate hierarchies with color-coded boxes for value nodes are produced for all rows of u. Typically. .conversion. 1.aggregation. In the latter case.conversion.8). NA. "nodes"). This input is only considered if the argument type is specified to be either "hierarchy" or "table". this will be the complete output or an output row of the function evaluate. 1. It is then used to color-code the boxes of the hierarchy representing value nodes or the table.attrib .utility.conversion. TRUE. this will be the complete output or an output row of the function class ’utility. then the median and quantile boxes are plotted for value nodes or the expected utility for utility nodes unless the argument main contains as many elements as the number of rows of u. utility. .2. If u is a data frame with more than one row and the argument type is equal to "hierarchy". c("hierarchy". u uref par type nodes col gridlines main cex.parfun plot.nodes with.calc.30 plot. (optional) vector or data frame with elements or columns labelled according to the nodes of the hierarchy containing values or utilities. (optional) vector or data frame with elements or columns labelled according to the nodes of the hierarchy containing values or utilities.. "". NA. "table".reaches f. "node". 1.parfun Plot Node Definition or Underlying Objectives Hierarchy Description Plot node definition or underlying objectives hierarchy.colors().) Arguments x u node to be plotted.parfun’ = = = = = = = = = = = = = = NA.4.2. uref . c( .main cex. NA.

(optional) character vector specifying the nodes for which the definitions will be plotted or which will be considered in a table.nodes .nodes cex.utility.parfun 31 evaluate. (optional) numeric vector of levels at which gridlines are plotted in node definitions. (optional) scaling factor for attribute labels used in the plot.conversion. type nodes col gridlines main cex. It is then used to color-code the upper part of the boxes of the hierarchy to allow for a comparison with the results provided by the argument u which are shown in the lower part of the boxes. the parameters have to be passed to evaluate.utility. (optional) fraction of the width of the plot reserved for the row labels of the table if the argument type is equal to "table". To color-code hierarchies or tables for different parameter values. This attribute is only used for value nodes and if values are provided by the arguments u and/or uref. (optional) title(s) of the plot. This input is only considered if the argument type is specified to be "hierarchy".aggregation. If the argument type is equal to "hierarchy" and the a vector of titles with the same length as the number of rows of the argument u is provided. (optional) scaling factor for title of the plot. (optional) specifies the type of plot to be produced. "nodes": produces plots of node definitions for all nodes defined by the attribute nodes. par (optional) labelled numeric parameter vector providing parameters to modify the value or utility function before plotting the node.aggregation before passing the results of this function to this plotting routine.utility.main cex. Options: "hierarchy". the medians and colored boxes indicating 90% credibility or occurrence ranges are plotted at all nodes. "node" or "nodes". The default value of NA indicates that all nodes will be plotted. (optional) character vector of colors to be used to color the interval between zero and unity in equidistant sections (use repetitions of the same color if you want to have a non-equidistant color-coding). "hierarchy": produces a plot of the objectives hierarchy including color-coded results for values or utilities if these values are provided by the arguments u and/or uref. Otherwise.plot. Note that this affects only the node definitions plotted if the argument type is specified to be "node" or "nodes". "table": produces a table with color-coded results for values or utilities if these values are provided by the argument u. This attribute is only used if the argument type is specified to be either "node" or "nodes". "node": produces a plot of the definition of the current node.reaches f. a color-coded hierarchy is plotted for each row of u. (optional) scaling factor for node labels used in the plot. "table". This argument only affects the output if the argument type was indicated to be either "table" or "nodes". (optional) fraction of the height of the plot reserved for the column labels of the table if the argument type is equal to "table".attrib f.

evaluating and visualizing value and utility functions for decision support.. S. 1993.. Note plot. H.parfun for how to evaluate the node. Textbooks on the use of utility and value functions in decision analysis: Keeney.intpol plot. See utility.colors for an example of how to construct color schemes and utility. Environmental Modelling & Software. Schuwirth. and Langhans. 2010.utility. Eisenfuehr.conversion. Weber. Springer.calc.conversion. N.create for how to construct such a node and evaluate. . additional arguments passed to the R plotting routine. F. and Raiffa. Examples # see help(utility) # for examples. P.attrib . Berlin. Cambridge University Press. T. First published in 1976. R.. See Also See utility. Note that the plotting routines plot.colors for how to get colors for specifed value levels. and Langer.parfun (optional) indicates if attributes should be listed if the argument type is equal to "hierarchy".conversion.utility.reichert@eawag. Rational Decision Making..get. Author(s) Peter Reichert <peter.. Constructing. L..parfun.32 with.conversion. in press.utility. Decisions with Multiple Objectives. M.aggregation are exactly the same so that all hierarchies can be plotted with exactly the same commands irrespective of the type of the top-level node.utility.ch> References Short description of the package: Reichert.

gridlines main cex.6.2.calc. c( ..endnode.cond 33 plot.parfun1d plot.endnode. (optional) labelled numeric parameter vector providing parameters to modify the value or utility function before plotting the node.cond Plot Node Definition Description Plot node definition. (optional) title of the plot.endnode. (optional) character vector specifying the names of the nodes to be plotted.main = nodes = .colors().endnode.intpol1d plot.4.utility.endnode.. (optional) numeric vector of levels at which gridlines are plotted in the node definition.endnode.) Arguments x par col node to be plotted.endnode.discrete plot.8).utility..utility. (optional) character vector of colors to be used to color the interval between zero and unity in equidistant sections (use repetitions of the same color if you want to have a non-equidistant color-coding). Usage ## S3 method for plot(x..utility.cond’ NA. par = col = gridlines = main = cex.utility.plot. x$name. "". (optional) scaling factor for title of the plot. class ’utility. This attribute is only used for value nodes.intpol2d are as far as possible the same so that all end nodes can be plotted with the same commands irrespective of the type of the end node. . . utility. Note Note that the plotting routines for the other end nodes plot. . additional arguments passed to the R plotting routine.main nodes . . 1.utility.

Decisions with Multiple Objectives. Rational Decision Making. .create for how to construct such a node and evaluate. See utility. M. and Raiffa.calc.main = . and Langer.discrete’ NA.cond. 1993.colors(). Berlin. L. R.utility. in press.2. and Langhans. . Usage ## S3 method for plot(x.utility.endnode.. utility. S.discrete Plot Node Definition Description Plot node definition. See Also See utility.) class ’utility.. .endnode. plot.8).. Textbooks on the use of utility and value functions in decision analysis: Keeney. ..get. Eisenfuehr.. Environmental Modelling & Software. 2010.colors for how to get colors for specifed value levels. 1.. c( .utility. Cambridge University Press.ch> References Short description of the package: plot. First published in 1976.endnode.discrete Reichert.34 Author(s) Peter Reichert <peter. Examples # see help(utility) # for examples.6.cond for how to evaluate the node.colors for an example of how to construct color schemes and utility. T. evaluating and visualizing value and utility functions for decision support. F. H. par = col = gridlines = main = cex.endnode. N. Constructing. P. Springer.calc.reichert@eawag. "".endnode.4. Schuwirth. Weber.

(optional) scaling factor for title of the plot. Eisenfuehr. 2010.main .endnode.utility. P. T.. M. Rational Decision Making. Springer.endnode.intpol2d plot.utility. 1993. L. F. Textbooks on the use of utility and value functions in decision analysis: Keeney. and Langhans. Decisions with Multiple Objectives.parfun1d plot.discrete Arguments x par col node to be plotted. evaluating and visualizing value and utility functions for decision support..plot. in press.endnode. additional arguments passed to the R plotting routine. Environmental Modelling & Software. Weber.create for how to construct such a node and evaluate.reichert@eawag.utility.. and Raiffa. (optional) title of the plot. This attribute is only used for value nodes. gridlines main cex. Constructing.discrete for how to evaluate the node. (optional) numeric vector of levels at which gridlines are plotted in the node definition. (optional) character vector of colors to be used to color the interval between zero and unity in equidistant sections (use repetitions of the same color if you want to have a non-equidistant color-coding). . Schuwirth.intpol1d plot. Berlin.utility.cond are as far as possible the same so that all end nodes can be plotted with the same commands irrespective of the type of the end node. H.utility.ch> References Short description of the package: Reichert. Cambridge University Press. R...endnode. 35 (optional) labelled numeric parameter vector providing parameters to modify the value or utility function before plotting the node.utility. N. Author(s) Peter Reichert <peter. See Also See utility.discrete. and Langer..endnode. S. Note Note that the plotting routines for the other end nodes plot. First published in 1976.endnode.endnode.

(optional) character vector of colors to be used to color the interval between zero and unity in equidistant sections (use repetitions of the same color if you want to have a non-equidistant color-coding).calc. class ’utility. (optional) title of the plot.36 plot.) Arguments x par col node to be plotted.endnode. (optional) labelled numeric parameter vector providing parameters to modify the value or utility function before plotting the node. .intpol1d See utility. "".calc. par = col = gridlines = main = cex.2. .colors for an example of how to construct color schemes and utility. (optional) scaling factor for title of the plot..6.utility.endnode. . Usage ## S3 method for plot(x. .8). 1.intpol1d Plot Node Definition Description Plot node definition. additional arguments passed to the R plotting routine. (optional) numeric vector of levels at which gridlines are plotted in the node definition.get. gridlines main cex.main = . utility. Examples # see help(utility) # for examples. c( .colors for how to get colors for specifed value levels.endnode.main . This attribute is only used for value nodes.. plot..utility.4.colors()..intpol1d’ NA.

endnode.utility.intpol2d Note 37 Note that the plotting routines for the other end nodes plot. and Langhans. plot. P. Textbooks on the use of utility and value functions in decision analysis: Keeney. First published in 1976.utility.endnode.plot. 1993.utility. Berlin.discrete plot.utility.parfun1d plot.colors for an example of how to construct color schemes and utility. M. Examples # see help(utility) # for examples. Constructing.endnode..endnode. H.get. R.calc. and Langer.intpol2d Plot Node Definition Description Plot node definition.endnode.endnode. Decisions with Multiple Objectives. Schuwirth. N. Environmental Modelling & Software. and Raiffa. See utility.endnode. 2010. Rational Decision Making.utility.intpol1d for how to evaluate the node. See Also See utility. T.. Weber.ch> References Short description of the package: Reichert. F.utility. Cambridge University Press..create for how to construct such a node and evaluate.colors for how to get colors for specifed value levels.reichert@eawag.endnode. S.intpol1d. Springer.. .utility. in press. evaluating and visualizing value and utility functions for decision support. Eisenfuehr.cond are as far as possible the same so that all end nodes can be plotted with the same commands irrespective of the type of the end node.intpol2d plot. L. Author(s) Peter Reichert <peter.

evaluating and visualizing value and utility functions for decision support. par = col = gridlines = main = cex. class ’utility. plot. (optional) title of the plot.discrete plot.endnode.reichert@eawag. (optional) character vector of colors to be used to color the interval between zero and unity in equidistant sections (use repetitions of the same color if you want to have a non-equidistant color-coding). additional arguments passed to the R plotting routine.main = . Environmental Modelling & Software. Author(s) Peter Reichert <peter.endnode.utility.intpol2d (optional) labelled numeric parameter vector providing parameters to modify the value or utility function before plotting the node. (optional) numeric vector of levels at which gridlines are plotted in the node definition. N. utility.calc. (optional) scaling factor for title of the plot.6.. S. . .endnode. 1. .colors(). Schuwirth.endnode. Constructing. gridlines main cex.2.main .38 Usage ## S3 method for plot(x.ch> References Short description of the package: Reichert.4. Not used for this type of node..intpol2d plot. and Langhans. This attribute is only used for value nodes.parfun1d plot.utility.endnode.) Arguments x par col node to be plotted. in press. c( .8).utility..utility.. Textbooks on the use of utility and value functions in decision analysis: . Note Note that the plotting routines for the other end nodes plot..utility.cond are as far as possible the same so that all end nodes can be plotted with the same commands irrespective of the type of the end node.intpol2d’ NA.endnode. "". P..

intpol1d. See utility.endnode. . H. . Berlin. par = col = gridlines = main = cex. and Langer. T. . "".) Arguments x par col node to be plotted.utility. and Raiffa. c( .parfun1d’ NA. (optional) character vector of colors to be used to color the interval between zero and unity in equidistant sections (use repetitions of the same color if you want to have a non-equidistant color-coding).calc.colors(). Springer. Weber.colors for how to get colors for specifed value levels. This attribute is only used for value nodes.6.endnode.2. . R.endnode.endnode.get. plot. See Also See utility.main = . utility.. Examples # see help(utility) # for examples. Cambridge University Press..8).intpol1d for how to evaluate the node. M. 1993..parfun1d 39 Keeney. Decisions with Multiple Objectives. Eisenfuehr. class ’utility. 2010.utility.parfun1d Plot Node Definition Description Plot node definition. F.plot. Rational Decision Making.. 1.utility. Usage ## S3 method for plot(x.create for how to construct such a node and evaluate.4. First published in 1976. (optional) labelled numeric parameter vector providing parameters to modify the value or utility function before plotting the node.endnode.colors for an example of how to construct color schemes and utility. L.calc.

. Note that the plotting routines for the other end nodes plot. Springer. Examples # see help(utility) # for examples. L.endnode. Rational Decision Making. Textbooks on the use of utility and value functions in decision analysis: Keeney.colors for how to get colors for specifed value levels. and Langer.. Weber.parfun1d for how to evaluate the node.endnode. T. in press.endnode. 2010.discrete plot. Eisenfuehr. (optional) title of the plot.utility.utility. P..parfun1d (optional) numeric vector of levels at which gridlines are plotted in the node definition. and Langhans. H.intpol1d plot. Cambridge University Press. First published in 1976. Constructing.40 gridlines main cex.create for how to construct such a node and evaluate.endnode.cond are as far as possible the same so that all end nodes can be plotted with the same commands irrespective of the type of the end node. (optional) scaling factor for title of the plot.endnode.utility. Decisions with Multiple Objectives.endnode.utility. Note plot. 1993. evaluating and visualizing value and utility functions for decision support.. and Raiffa. N..parfun1d. Environmental Modelling & Software. Berlin.utility. additional arguments passed to the R plotting routine.endnode.colors for an example of how to construct color schemes and utility.. F.intpol2d plot.calc. M.reichert@eawag. S. R. Schuwirth..utility. See utility. See Also See utility.main . Author(s) Peter Reichert <peter.get.ch> References Short description of the package: Reichert.

. Springer.utility. and Langhans..print. evaluating and visualizing value and utility functions for decision support. P. currently no other arguments are implemented or passed further. . Cambridge University Press.. in press.utility. M. Eisenfuehr. N. First published in 1976. Environmental Modelling & Software. T.aggregation’ print(x. See Also See utility.aggregation 41 print. Decisions with Multiple Objectives. S.) Arguments x . 2010. F. and Langer. and Raiffa. Schuwirth.reichert@eawag. Berlin. Constructing. Author(s) Peter Reichert <peter. node to be printed.. H. R.aggregation Print Definitions of Node and Associated Hierarchy Description Print definition of node and associated hierarchy. 1993.. ..create for how to construct such a node. L.aggregation.. Note In the current version of the package.ch> References Short description of the package: Reichert. Rational Decision Making.. Textbooks on the use of utility and value functions in decision analysis: Keeney. Weber. the methods print and summary provide the same output. Usage ## S3 method for class ’utility.

Author(s) Peter Reichert <peter. Springer.intpol.intpol Print Definitions of Node and Associated Hierarchy Description Print definition of node and associated hierarchy.ch> References Short description of the package: Reichert. print. in press. Environmental Modelling & Software. H.intpol’ print(x. T. P. Eisenfuehr.. Schuwirth.. First published in 1976. and Langhans.42 Examples # see help(utility) # for examples. and Langer.. evaluating and visualizing value and utility functions for decision support. 1993.utility. 2010. See Also See utility. . Cambridge University Press. F. Berlin. . Decisions with Multiple Objectives.. L. N.intpol print. currently no other arguments are implemented or passed further. Constructing. Textbooks on the use of utility and value functions in decision analysis: Keeney.conversion.create for how to construct such a node.conversion. node to be printed.conversion.. Rational Decision Making. the methods print and summary provide the same output.reichert@eawag. Usage ## S3 method for class ’utility. R.. M..utility.conversion. Note In the current version of the package. S. Weber. and Raiffa.) Arguments x ..

.parfun Print Definitions of Node and Associated Hierarchy Description Print definition of node and associated hierarchy. Textbooks on the use of utility and value functions in decision analysis: Keeney.create for how to construct such a node. First published in 1976. Decisions with Multiple Objectives. and Langer.. F.ch> References Short description of the package: Reichert..conversion.parfun’ print(x. and Raiffa.) Arguments x ..conversion.reichert@eawag. Weber.print. Constructing. 1993. Schuwirth. L.parfun. See Also See utility. Note In the current version of the package. H.conversion.conversion.. . 2010. R..utility. Usage ## S3 method for class ’utility. and Langhans. Author(s) Peter Reichert <peter. the methods print and summary provide the same output.utility. T.. Berlin. evaluating and visualizing value and utility functions for decision support. 43 print. S. in press. P. node to be printed. .. currently no other arguments are implemented or passed further. Rational Decision Making. N. Eisenfuehr. M. Environmental Modelling & Software. Cambridge University Press. Springer.parfun Examples # see help(utility) # for examples.

1993. . Decisions with Multiple Objectives. currently no other arguments are implemented or passed further. Environmental Modelling & Software. Usage ## S3 method for class ’utility.. N.endnode.reichert@eawag.cond print. the methods print and summary provide the same output..44 Examples # see help(utility) # for examples. Springer.ch> References Short description of the package: Reichert. Cambridge University Press. Textbooks on the use of utility and value functions in decision analysis: Keeney. node to be printed. F.endnode. Author(s) Peter Reichert <peter... M.. Constructing.endnode. S. and Langhans. Schuwirth.utility. . See Also See utility.cond Print Node Definition Description Print node defintion. L. and Raiffa. Eisenfuehr.utility.cond’ print(x. T. 2010..) Arguments x .cond. R.create for how to construct such a node. Note In the current version of the package. Berlin. print. First published in 1976. Rational Decision Making. H.. Weber. and Langer. in press. P.endnode.. evaluating and visualizing value and utility functions for decision support.

First published in 1976.discrete Print Node Definition Description Print node defintion.. and Langer.endnode. S. 45 print.endnode. See Also See utility. Note In the current version of the package. Author(s) Peter Reichert <peter. Schuwirth.. Rational Decision Making. in press. 2010. L. Usage ## S3 method for class ’utility. Textbooks on the use of utility and value functions in decision analysis: Keeney. node to be printed. Weber. and Langhans.print. N.create for how to construct such a node. evaluating and visualizing value and utility functions for decision support. Berlin.. Eisenfuehr.endnode. Springer. . Constructing. R.endnode. the methods print and summary provide the same output... Environmental Modelling & Software. . F.. T. Decisions with Multiple Objectives.utility. and Raiffa..ch> References Short description of the package: Reichert. M. currently no other arguments are implemented or passed further. H.discrete’ print(x. Cambridge University Press.) Arguments x .reichert@eawag.. 1993.utility. P.discrete Examples # see help(utility) # for examples.discrete.

F..intpol1d. Eisenfuehr. Weber. Schuwirth. and Langhans.. Springer..utility. P. Note In the current version of the package. Cambridge University Press. First published in 1976. . S. Environmental Modelling & Software. in press. and Langer. Berlin. See Also See utility. 2010. L..) Arguments x .46 Examples # see help(utility) # for examples. currently no other arguments are implemented or passed further.. Rational Decision Making. evaluating and visualizing value and utility functions for decision support.intpol1d print.intpol1d Print Node Definition Description Print node defintion. Constructing. Textbooks on the use of utility and value functions in decision analysis: Keeney.reichert@eawag. . 1993. M. node to be printed.. N. Usage ## S3 method for class ’utility. the methods print and summary provide the same output.. print.endnode.. Decisions with Multiple Objectives. Author(s) Peter Reichert <peter.endnode. and Raiffa. T.utility.intpol1d’ print(x.ch> References Short description of the package: Reichert.create for how to construct such a node. R.endnode. H.endnode.

reichert@eawag.endnode.endnode..endnode. Textbooks on the use of utility and value functions in decision analysis: Keeney.endnode... H. M. and Raiffa. 2010. Decisions with Multiple Objectives.intpol2d Print Node Definition Description Print node defintion. S. 47 print. in press. R. L. Rational Decision Making. Usage ## S3 method for class ’utility. Schuwirth.ch> References Short description of the package: Reichert. evaluating and visualizing value and utility functions for decision support. Cambridge University Press.print.. Eisenfuehr.utility. and Langer. Constructing. 1993. .utility. Berlin.intpol2d Examples # see help(utility) # for examples.intpol2d’ print(x.. Author(s) Peter Reichert <peter. Note In the current version of the package. N. and Langhans.create for how to construct such a node.. . node to be printed. P. First published in 1976. See Also See utility.) Arguments x . Springer. the methods print and summary provide the same output. Environmental Modelling & Software. currently no other arguments are implemented or passed further. T.. F. Weber..intpol2d.

. Textbooks on the use of utility and value functions in decision analysis: Keeney. Weber. H. and Langer. F. Note In the current version of the package. Rational Decision Making. print.create for how to construct such a node.ch> References Short description of the package: Reichert.utility.. Constructing. the methods print and summary provide the same output. ..endnode.parfun1d print.) Arguments x . P.reichert@eawag. M. L. 2010. evaluating and visualizing value and utility functions for decision support. and Raiffa. R.48 Examples # see help(utility) # for examples.endnode.. Author(s) Peter Reichert <peter..endnode. 1993. Eisenfuehr. T. First published in 1976. and Langhans. S. Schuwirth. Environmental Modelling & Software.. Usage ## S3 method for class ’utility.parfun1d Print Node Definition Description Print node defintion. .parfun1d’ print(x. in press.endnode.. Decisions with Multiple Objectives. currently no other arguments are implemented or passed further. N. node to be printed. Berlin..utility. See Also See utility.parfun1d. Cambridge University Press. Springer.

Environmental Modelling & Software. 1993. Berlin. the methods print and summary provide the same output. Decisions with Multiple Objectives. evaluating and visualizing value and utility functions for decision support.) Arguments object ... See Also See utility.reichert@eawag. P.create for how to construct such a node.. N. and Langer.aggregation. node of which a summary is to be printed. Springer..utility. Eisenfuehr. 49 summary. First published in 1976. and Langhans.aggregation’ summary(object.ch> References Short description of the package: Reichert.aggregation Examples # see help(utility) # for examples. currently no other arguments are implemented or passed further. Note In the current version of the package. and Raiffa. T. S. R. Usage ## S3 method for class ’utility.summary.. . L. M. Constructing. Weber. Textbooks on the use of utility and value functions in decision analysis: Keeney. Author(s) Peter Reichert <peter. Cambridge University Press. Rational Decision Making. . in press.utility. F. Schuwirth.. H.aggregation Print Summary of Definitions of Node and Associated Hierarchy Description Print summary of definition of node and associated hierarchy... 2010.

2010.. the methods print and summary provide the same output.50 Examples # see help(utility) # for examples..intpol’ summary(object.conversion. First published in 1976..) Arguments object . Rational Decision Making. in press. . H. S. Environmental Modelling & Software. Textbooks on the use of utility and value functions in decision analysis: Keeney. Decisions with Multiple Objectives. Schuwirth.reichert@eawag.intpol. currently no other arguments are implemented or passed further. L. R. node of which a summary is to be printed. and Langer.conversion. and Langhans. See Also See utility..conversion.. N. T.utility. summary.utility. Weber. F. Usage ## S3 method for class ’utility. Cambridge University Press. Springer. . 1993.intpol Print Summary of Definitions of Node and Associated Hierarchy Description Print summary of definition of node and associated hierarchy. evaluating and visualizing value and utility functions for decision support.create for how to construct such a node.. P.conversion. Note In the current version of the package. Author(s) Peter Reichert <peter. Berlin. Eisenfuehr. M.intpol summary. and Raiffa...ch> References Short description of the package: Reichert. Constructing.

N. in press. node of which a summary is to be printed.conversion. currently no other arguments are implemented or passed further. L. the methods print and summary provide the same output.. Usage ## S3 method for class ’utility. Springer. and Raiffa. Decisions with Multiple Objectives. Berlin.conversion. Author(s) Peter Reichert <peter. Constructing. S. 51 summary. Note In the current version of the package. Weber.utility.reichert@eawag..parfun Print Summary of Definitions of Node and Associated Hierarchy Description Print summary of definition of node and associated hierarchy.ch> References Short description of the package: Reichert. F.utility. . Eisenfuehr. M. H.summary. See Also See utility. Schuwirth.create for how to construct such a node...) Arguments object .parfun. Textbooks on the use of utility and value functions in decision analysis: Keeney.. Rational Decision Making. R.conversion. Cambridge University Press. First published in 1976.. and Langhans. Environmental Modelling & Software.. 2010. P. evaluating and visualizing value and utility functions for decision support.conversion.parfun’ summary(object. T.parfun Examples # see help(utility) # for examples.. and Langer. 1993. .

N. Textbooks on the use of utility and value functions in decision analysis: Keeney. Constructing. and Raiffa.endnode.. Environmental Modelling & Software. H.create for how to construct such a node. and Langhans.endnode.cond’ summary(object. S. R. Weber. Note In the current version of the package. node of which a summary is to be printed.52 Examples # see help(utility) # for examples. the methods print and summary provide the same output.cond Print Summary of Node Definition Description Print summary of node definition. F. summary. . Springer. M. 2010.ch> References Short description of the package: Reichert..reichert@eawag. 1993.. Schuwirth. Cambridge University Press. P.... evaluating and visualizing value and utility functions for decision support. First published in 1976. See Also See utility. T. in press.endnode. L.. Decisions with Multiple Objectives.cond. Eisenfuehr.utility. .endnode. Usage ## S3 method for class ’utility.utility.. Rational Decision Making.) Arguments object . Author(s) Peter Reichert <peter. Berlin. currently no other arguments are implemented or passed further.cond summary. and Langer.

.utility. N.ch> References Short description of the package: Reichert.discrete Examples # see help(utility) # for examples. H. Berlin. .summary. Note In the current version of the package. See Also See utility. 1993.endnode. the methods print and summary provide the same output. Decisions with Multiple Objectives. currently no other arguments are implemented or passed further..utility. Environmental Modelling & Software. Usage ## S3 method for class ’utility... S. F. First published in 1976. Cambridge University Press.) Arguments object . P. and Langhans.. 53 summary.endnode. Springer. R. Schuwirth. Weber. M.discrete.discrete Print Summary of Node Definition Description Print summary of node definition. Author(s) Peter Reichert <peter. T.discrete’ summary(object.. L.endnode.endnode..reichert@eawag. node of which a summary is to be printed.create for how to construct such a node. Textbooks on the use of utility and value functions in decision analysis: Keeney. and Langer. Eisenfuehr.. in press. 2010.. Rational Decision Making. Constructing. evaluating and visualizing value and utility functions for decision support. and Raiffa.

. summary.54 Examples # see help(utility) # for examples. R. Environmental Modelling & Software. Author(s) Peter Reichert <peter. Springer..reichert@eawag. M.endnode. S. the methods print and summary provide the same output. 1993. T.utility. First published in 1976. evaluating and visualizing value and utility functions for decision support... Schuwirth. Constructing.endnode. 2010. and Langer. and Raiffa. Textbooks on the use of utility and value functions in decision analysis: Keeney. Eisenfuehr. ..ch> References Short description of the package: Reichert.intpol1d.utility.) Arguments object . H. . Weber. Berlin. Cambridge University Press.. N. currently no other arguments are implemented or passed further. P.endnode.endnode. in press. Note In the current version of the package.intpol1d summary. and Langhans. Rational Decision Making..intpol1d’ summary(object. F. See Also See utility. L. Usage ## S3 method for class ’utility.create for how to construct such a node. Decisions with Multiple Objectives.. node of which a summary is to be printed.intpol1d Print Summary of Node Definition Description Print summary of node definition.

utility. P. Schuwirth.utility. F. T. Berlin. Eisenfuehr. and Raiffa. Weber. evaluating and visualizing value and utility functions for decision support. Rational Decision Making.. L.) Arguments object .endnode.endnode.endnode. Springer. Author(s) Peter Reichert <peter. node of which a summary is to be printed.intpol2d Print Summary of Node Definition Description Print summary of node definition. See Also See utility. the methods print and summary provide the same output.ch> References Short description of the package: Reichert.summary. Textbooks on the use of utility and value functions in decision analysis: Keeney. Environmental Modelling & Software. . 2010. S. Decisions with Multiple Objectives. Cambridge University Press. Note In the current version of the package.intpol2d Examples # see help(utility) # for examples. First published in 1976.. N.intpol2d. in press. 55 summary.reichert@eawag. and Langer.intpol2d’ summary(object. 1993..create for how to construct such a node. and Langhans. ..endnode.. H. Constructing. currently no other arguments are implemented or passed further.. R. M.. Usage ## S3 method for class ’utility..

and Raiffa. Eisenfuehr.utility. Usage ## S3 method for class ’utility. L.56 Examples # see help(utility) # for examples. Springer.parfun1d summary.ch> References Short description of the package: Reichert.parfun1d Print Summary of Node Definition Description Print summary of node definition..endnode. Environmental Modelling & Software.. 1993. the methods print and summary provide the same output. T.. node of which a summary is to be printed.reichert@eawag. Schuwirth.. M.endnode.endnode. R. Berlin. and Langer. Note In the current version of the package. Author(s) Peter Reichert <peter. See Also See utility. P... summary.. H.create for how to construct such a node.parfun1d. F. Constructing. and Langhans. N. currently no other arguments are implemented or passed further. in press. Decisions with Multiple Objectives.) Arguments object . Cambridge University Press. evaluating and visualizing value and utility functions for decision support. First published in 1976.. .utility. .parfun1d’ summary(object. Textbooks on the use of utility and value functions in decision analysis: Keeney. Rational Decision Making. Weber. 2010. S.endnode.

. Schuwirth. First published in 1976. and Langhans. F. in press. Constructing. parameter values can be provided by an additional argument par.updatepar Examples # see help(utility) # for examples. . and Langer. Eisenfuehr. Textbooks on the use of utility and value functions in decision analysis: Keeney. S. Author(s) Peter Reichert <peter.) Arguments x . T.. evaluating and visualizing value and utility functions for decision support...reichert@eawag. Decisions with Multiple Objectives. Usage updatepar(x.ch> References Short description of the package: Reichert. H.. Springer.. Environmental Modelling & Software. Value The node or node hierarchy with updated parameters is returned. L. and Raiffa. Cambridge University Press.. Weber. R. Rational Decision Making. M. 1993. 2010. 57 updatepar Update Parameters in Node Definitions Description Generic function to update parameters in all node defintions of the hierarchy defined by the given node. N. Berlin. node to be updated.. P..

parfun1d updatepar.utility.utility.parfun1d.intpol updatepar.endnode. utility.endnode. updatepar.conversion.utility.create.cond updatepar.58 See Also See utility..parfun.endnode.aggregation’ updatepar(x.) Arguments x par . utility.endnode.endnode.endnode. utility.intpol2d updatepar.create.discrete updatepar.create..endnode.utility.intpol2d.create for how to construct the nodes and updatepar. parameter vector with labelled parameters to be updated. utility. . currently no other arguments are implemented or passed further. Usage ## S3 method for class ’utility.conversion.utility.aggregation updatepar.intpol1d updatepar.utility.create.aggregation Update Parameters in Node Definitions Description Update parameters in all node defintions of the hierarchy defined by the node.utility.aggregation updatepar.cond. utility.create.endnode.intpol.utility. utility.utility.endnode. node to be updated.create. ..discrete.endnode.. utility.parfun for the updates of the specific nodes.intpol1d. Value The node hierarchy with updated parameters is returned.conversion.conversion.create.utility.aggregation. par=NA.

Decisions with Multiple Objectives. P.intpol Update Parameters in Node Definitions Description Update parameters in all node defintions of the hierarchy defined by the node. Weber. Springer. in press. M. Schuwirth. Cambridge University Press.endnode. First published in 1976. . and Langer. currently no other arguments are implemented or passed further. T.endnode. See Also See utility.cond updatepar.conversion.ch> References Short description of the package: 59 Reichert.utility.parfun for analogous updates of other nodes updatepar.. N. 2010.utility. par=NA. F.conversion. L.reichert@eawag.create for how to construct such a node and updatepar.endnode. and Raiffa.conversion.endnode..aggregation. Environmental Modelling & Software. Textbooks on the use of utility and value functions in decision analysis: Keeney. Constructing.conversion.intpol’ updatepar(x.utility. and Langhans.) Arguments x par .utility..endnode. parameter vector labelled with parameter values to be updated.. Usage ## S3 method for class ’utility.utility.intpol updatepar.intpol1d updatepar.discrete updatepar. 1993..utility.conversion. node to be updated. Rational Decision Making..utility. H. S.. evaluating and visualizing value and utility functions for decision support.utility.. Eisenfuehr.updatepar.utility.intpol Author(s) Peter Reichert <peter. Berlin.parfun1d updatepar. R.intpol2d updatepar. .

conversion. in press. First published in 1976.utility. P.parfun1d.conversion.utility. Springer. Eisenfuehr.create for how to construct such a node and updatepar. and Raiffa.utility.utility. R.utility.cond updatepar. .parfun Update Parameters in Node Definitions Description Update parameters in all node defintions of the hierarchy defined by the node.. Berlin.discrete updatepar.conversion.reichert@eawag. par=NA. N. 2010.intpol2d updatepar. See Also See utility. Cambridge University Press...endnode. M. Usage ## S3 method for class ’utility. Environmental Modelling & Software.conversion..utility. Author(s) Peter Reichert <peter.ch> References Short description of the package: updatepar. T. Textbooks on the use of utility and value functions in decision analysis: Keeney.parfun1d updatepar.parfun for analogous updates of other nodes updatepar. and Langer.intpol1d updatepar.parfun’ updatepar(x. Rational Decision Making. L.aggregation updatepar. F. H. 1993.endnode. S. Constructing..endnode. Decisions with Multiple Objectives..endnode.utility. evaluating and visualizing value and utility functions for decision support.parfun Reichert.60 Value The node hierarchy with updated parameters is returned.utility.endnode.endnode.utility. and Langhans.) . Schuwirth. Weber.

and Raiffa.intpol2d updatepar. Textbooks on the use of utility and value functions in decision analysis: Keeney. Environmental Modelling & Software.discrete updatepar.parfun1d.cond Update Parameters in Node Definitions Description Update parameters in all node defintions used to define the node. 2010.ch> References Short description of the package: node to be updated.intpol for analogous updates of other nodes updatepar.endnode. parameter vector with labelled parameters to be updated.. F.parfun1d updatepar.aggregation updatepar. L.updatepar.utility. and Langer. Value The node hierarchy with updated parameters is returned.endnode. See Also See utility.utility.utility. Berlin.endnode.cond updatepar.endnode.endnode. H.. in press.conversion. M.endnode. Constructing. P. S.intpol1d updatepar. Rational Decision Making. currently no other arguments are implemented or passed further. N. Springer.create for how to construct such a node and updatepar. and Langhans.utility. R.utility..utility.reichert@eawag.. First published in 1976.utility.endnode. Cambridge University Press.. Weber.cond Arguments x par .endnode.. 1993. T.utility. 61 Reichert. Author(s) Peter Reichert <peter.utility. evaluating and visualizing value and utility functions for decision support. Schuwirth. Decisions with Multiple Objectives. Eisenfuehr. .

L.. .endnode.intpol2d updatepar..ch> References Short description of the package: node to be updated.parfun1d.cond’ updatepar(x. Rational Decision Making.intpol1d updatepar..endnode. P. Eisenfuehr.utility. First published in 1976. F. and Langer.intpol updatepar.aggregation updatepar.discrete updatepar.. Weber.62 Usage ## S3 method for class ’utility.cond parameter vector with labelled parameters to be updated..endnode. S.parfun1d updatepar. Environmental Modelling & Software.endnode.utility. Author(s) Peter Reichert <peter. T. in press. and Langhans. evaluating and visualizing value and utility functions for decision support. R. Textbooks on the use of utility and value functions in decision analysis: Keeney.endnode..utility.utility.utility.utility. Decisions with Multiple Objectives..parfun for analogous updates of other nodes . See Also See utility. M.utility. Reichert.. H. N.conversion. Cambridge University Press.create for how to construct such a node and updatepar. Springer. Constructing.utility. 1993. currently no other arguments are implemented or passed further. Value The node with updated parameters is returned. updatepar.endnode.endnode.) Arguments x par .reichert@eawag. 2010.conversion. par=NA. and Raiffa. Berlin. Schuwirth.

Usage ## S3 method for class ’utility. .. S.parfun1d node to be updated. Textbooks on the use of utility and value functions in decision analysis: Keeney.ch> References Short description of the package: Reichert. Rational Decision Making.endnode. See Also See utility. T..discrete Update Parameters in Node Definition Description Update parameters in node defintion.utility..create for how to construct such a node and updatepar. Springer. N. currently no other arguments are implemented or passed further. P.) Arguments x par .utility.endnode.updatepar. Eisenfuehr..reichert@eawag.utility. First published in 1976.utility. Environmental Modelling & Software. H.. Constructing..discrete’ updatepar(x. . parameter vector with labelled parameters to be updated. evaluating and visualizing value and utility functions for decision support. and Raiffa. 2010. Decisions with Multiple Objectives. and Langhans..intpol1d updatepar. Value The node with updated parameters is returned. Weber.discrete 63 updatepar.. F.endnode. in press. L. par=NA.endnode.endnode.endnode. 1993.parfun1d. R. Schuwirth. Author(s) Peter Reichert <peter. and Langer. M. Cambridge University Press. Berlin.

Weber. currently no other arguments are implemented or passed further. Eisenfuehr. Springer. Textbooks on the use of utility and value functions in decision analysis: Keeney.utility..ch> References Short description of the package: Reichert.endnode.utility. H.... par=NA..utility. .intpol2d updatepar. First published in 1976.conversion. 1993. and Langer. Usage ## S3 method for class ’utility.intpol updatepar. N. Value The node with updated parameters is returned.endnode.endnode.intpol1d Update Parameters in Node Definition Description Update parameters in node defintion.64 updatepar. S. .utility.parfun for analogous updates of other nodes updatepar. in press. Cambridge University Press.reichert@eawag.endnode. parameter vector with labelled parameters to be updated. M.utility. Schuwirth. Environmental Modelling & Software.. Berlin.intpol1d’ updatepar(x. T.endnode.utility. P.. Constructing. L. Author(s) Peter Reichert <peter.conversion. Decisions with Multiple Objectives. R. F. evaluating and visualizing value and utility functions for decision support. and Raiffa. 2010. and Langhans.cond updatepar.aggregation updatepar. Rational Decision Making.intpol1d updatepar.) Arguments x par .. node to be updated.utility.

parfun1d. Value The node with updated parameters is returned.) Arguments x par .utility.endnode.endnode. . par=NA..intpol2d’ updatepar(x.intpol updatepar.intpol2d See Also See utility.endnode..utility.endnode.utility. currently no other arguments are implemented or passed further.updatepar.utility.intpol2d updatepar.endnode.ch> References Short description of the package: Reichert.utility. in press.utility.conversion. and Langhans.endnode. parameter vector with labelled parameters to be updated. Textbooks on the use of utility and value functions in decision analysis: node to be updated.create for how to construct such a node and updatepar.reichert@eawag. Schuwirth. Author(s) Peter Reichert <peter. evaluating and visualizing value and utility functions for decision support.utility. N.discrete updatepar.utility.intpol2d Update Parameters in Node Definition Description Update parameters in node defintion. Environmental Modelling & Software. P...parfun1d updatepar.aggregation updatepar. Usage ## S3 method for class ’utility. .parfun for analogous updates of other nodes 65 updatepar. S.conversion.cond updatepar. Constructing.endnode..endnode.utility..

2010. Berlin.) Arguments x par .cond updatepar. Value The node with updated parameters is returned.aggregation updatepar..endnode.reichert@eawag. See Also See utility.66 updatepar. 1993. par=NA.endnode.parfun1d Keeney.intpol updatepar.utility.discrete updatepar. Springer.. and Langer. .endnode.. Cambridge University Press..endnode.utility. R. Author(s) Peter Reichert <peter. M.utility.conversion. F. parameter vector with labelled parameters to be updated.utility. H. First published in 1976.utility. Usage ## S3 method for class ’utility. T.parfun1d Update Parameters in Node Definition Description Update parameters in node defintion. Decisions with Multiple Objectives..intpol1d updatepar.parfun1d updatepar.endnode.endnode.parfun1d..utility.utility.parfun1d’ updatepar(x. .create for how to construct such a node and updatepar. Rational Decision Making.conversion.ch> node to be updated.endnode.utility. and Raiffa.endnode.parfun for analogous updates of other nodes updatepar. L.utility. Weber. Eisenfuehr. currently no other arguments are implemented or passed further.

. Textbooks on the use of utility and value functions in decision analysis: Keeney.cond updatepar. In case of missing values in the vector u.conversion. N.utility. Environmental Modelling & Software.intpol updatepar. Decisions with Multiple Objectives. Weber. Rational Decision Making. L. Eisenfuehr.create for how to construct such a node and updatepar. and Langer. the weights of the non-missing components will be rescaled to sum to unity. Cambridge University Press.. P.add(u. 1993.utility. in press. R..endnode. Schuwirth.utility. and Raiffa.aggregate.. S. T.add Additive aggregation of values or utilities Description Function to perform an additive aggregation (weighted mean) of values or utilities. H.. Berlin.intpol2d updatepar. See Also See utility.endnode.endnode. F.parfun for analogous updates of other nodes utility. The weights need not be normalized.conversion. Springer. Constructing.aggregate.discrete updatepar. 2010. and Langhans.aggregation updatepar. evaluating and visualizing value and utility functions for decision support.add References Short description of the package: 67 Reichert.utility. M.aggregate. par) Arguments u par numeric vector of values or utilities to be aggregated.utility.intpol1d updatepar.utility.endnode. First published in 1976. numeric vector of weights for calculating the weighted mean of the values provided in the argument u. Usage utility.utility.utility.endnode. they will be normalized before use.parfun1d.

aggregate. and Raiffa.aggregate. Textbooks on the use of utility and value functions in decision analysis: Keeney.cobbdouglas. Springer. . Decisions with Multiple Objectives. Weber.max.aggregate.cobbdouglas Cobb-Douglas aggregation of values or utilities Description Function to perform a Cobb-Douglas aggregation (weighted geometric mean) of values or utilities.mult.. F. N. par=c(1. Constructing. Examples utility..reichert@eawag. Environmental Modelling & Software.ch> References Short description of the package: Reichert.aggregate. P. utility. utility.create Alternative aggregation techniques: utility. and Langhans.aggregation. T. Cambridge University Press.. H. Berlin. and Langer. utility. evaluating and visualizing value and utility functions for decision support. L. First published in 1976.aggregate. Rational Decision Making. 1993.8). R.aggregate. M.aggregate.add(c( . See Also Constructor of aggregation node: utility.min.cobbdouglas numeric value representing the weighted mean of the components of u. Schuwirth. S. Author(s) Peter Reichert <peter..68 Value utility. in press. 2010. utility.aggregate. Eisenfuehr.2.1)) utility.mix. .

and Raiffa. 1993.aggregate.cobbdouglas(u. Constructing. utility.aggregate.ch> References Short description of the package: Reichert.aggregate. they will be normalized before use.min. 2010.add..reichert@eawag. utility.aggregation. Springer. See Also Constructor of aggregation node: utility. Decisions with Multiple Objectives.aggregate. Textbooks on the use of utility and value functions in decision analysis: Keeney.aggregate. utility.create Alternative aggregation techniques: utility... Schuwirth. Eisenfuehr. H. evaluating and visualizing value and utility functions for decision support.. M.aggregate. Cambridge University Press.mult. R. F. par) Arguments u par numeric vector of values or utilities to be aggregated.utility. P. . Environmental Modelling & Software. the weights of the non-missing components will be rescaled to sum to unity. L. The weights need not be normalized. 69 numeric vector of weights for calculating the weighted geometric mean of the values provided in the argument u.cobbdouglas Usage utility.mix. First published in 1976. Author(s) Peter Reichert <peter. T. in press. In case of missing values in the vector u. and Langhans. S. Value The function returns the aggregated value or utility.aggregate. Berlin. N. utility.max. Rational Decision Making. Weber. and Langer.

and Langer. Constructing. Rational Decision Making.aggregate. Value maximum of the components of u.. T.ch> References Short description of the package: Reichert. 1993. R. Schuwirth.. Cambridge University Press. unused argument used for compatibility with other aggregation techniques that require parameters. Weber.8). Environmental Modelling & Software. First published in 1976. par=c(1.max(u. Textbooks on the use of utility and value functions in decision analysis: Keeney. F. S. . Eisenfuehr. L. H. M. 2010. par = NA) Arguments u par numeric vector of values or utilities to be aggregated. Berlin.70 Examples utility.1)) utility.cobbdouglas(c( . N.aggregate. .. and Langhans.max Maximum aggregation of values or utilities Description Function to perform a maximum aggregation of values or utilities..reichert@eawag.aggregate. and Raiffa. evaluating and visualizing value and utility functions for decision support. Author(s) Peter Reichert <peter. Decisions with Multiple Objectives.max utility.aggregate. Springer. in press.2. P. Usage utility.

aggregate. Usage utility.create Alternative aggregation techniques: utility.aggregate.aggregate. utility.aggregate. utility.min(u.utility.aggregate.aggregate. Value minimum of the components of u.aggregate. par = NA) Arguments u par numeric vector of values or utilities to be aggregated. Author(s) Peter Reichert <peter.8)) 71 utility. utility.mult.ch> .2.aggregate.max(c( .min See Also Constructor of aggregation node: utility.min.mix.add. unused argument used for compatibility with other aggregation techniques that require parameters. Examples utility.aggregate.reichert@eawag.cobbdouglas. .min Minimum aggregation of values or utilities Description Function to perform a minimum aggregation of values or utilities. utility.aggregation.

1993.add.cobbdouglas.mix(u.aggregate.2. Environmental Modelling & Software.aggregate.aggregate.mix Reichert.aggregate. in press.aggregation. Weber. R. See Also Constructor of aggregation node: utility. Decisions with Multiple Objectives. and Langhans.aggregate.. par) .aggregate. utility.mult. minimum and Cobb-Douglas aggregation techniques. P. utility. Usage utility.72 References Short description of the package: utility..aggregate.8)) utility.. F. Springer. Berlin. N..aggregate.min(c( . M. Textbooks on the use of utility and value functions in decision analysis: Keeney. First published in 1976.mix. S. The mixture consists of a weighted mean of the additive. H. T. Rational Decision Making. Cambridge University Press. and Langer. Eisenfuehr. Examples utility. utility.mix Mixed aggregation of values and utilities Description Function to perform a mixed aggregation of values and utilities.max.create Alternative aggregation techniques: utility. Constructing. and Raiffa. L. Schuwirth. evaluating and visualizing value and utility functions for decision support. .aggregate. 2010. utility.

aggregate. N.reichert@eawag. and Langhans.ch> References Short description of the package: Reichert.mult. .aggregate. and Langer. Decisions with Multiple Objectives..1)) . T. Environmental Modelling & Software. 1993.8).aggregate. the weights of the non-missing components will be rescaled to sum to unity.aggregate. 2010. Textbooks on the use of utility and value functions in decision analysis: Keeney.aggregation. L.par=c(1.mix Arguments u par numeric vector of values or utilities to be aggregated. Author(s) Peter Reichert <peter. .add.utility.max. Value The function returns the aggregated value or utility.min.create Alternative aggregation techniques: utility. First published in 1976.par=c(1. F. Rational Decision Making. R.8). Springer. utility.1.par=c(1. See Also Constructor of aggregation node: utility. )) utility. evaluating and visualizing value and utility functions for decision support.2.aggregate.mix(c( .aggregate. Berlin.aggregate. in press.. utility. H.2.mix(c( . . minimum and Cobb-Douglas aggregation techniques.1 .2. The weights need not be normalized.. 1.1 . Examples utility.aggregate. 73 numeric vector of weights for calculating the weighted mean of the values provided in the argument u followed by the three weights of the additive. M.mix(c( . utility. Cambridge University Press.. . Eisenfuehr. Weber. and Raiffa. they will be normalized before use. In case of missing values in the vector u.1 . Schuwirth.8). . )) utility. P. S.aggregate.cobbdouglas. Constructing. . . utility.

aggregate. Constructing. and Langer.aggregate. T. S.74 utility. F.aggregate.1..mult utility. and Langhans.. Value numeric value corresponding to the multiplicative aggregation of the values provided in the vector u. Berlin. ..par=c(1. 2010.mix(c( . L.ch> References Short description of the package: Reichert. Decisions with Multiple Objectives. Textbooks on the use of utility and value functions in decision analysis: Keeney.aggregate.1 . H. 1. Rational Decision Making. P. Eisenfuehr. 1993.8). numeric vector of weights for calculating the multiplicative combination of the values provided in the argument u. R.2. Usage utility. Schuwirth. Author(s) Peter Reichert <peter. Weber..mult Multiplicative aggregation of values or utilities Description Function to perform a multiplicative aggregation of values or utilities. N. par) Arguments u par numeric vector of values or utilities to be aggregated.reichert@eawag.mult(u. in press. and Raiffa. Environmental Modelling & Software. Springer. M. evaluating and visualizing value and utility functions for decision support. . Cambridge University Press.1)) utility. First published in 1976.

par=c( . This function must accept the arguments u and par which pass a vector of values or utilities to be aggregated and the parameters of the function. Usage utility.aggregation.required col shift.aggregate. par. utility. names.aggregation.par required num. Examples of functions provided by the package are = = = = = rep(NA. "black".aggregate. 1.create Alternative aggregation techniques: utility.3)) 75 utility. utility.aggregate.aggregate.create(name.cobbdouglas.add.aggregation.utility.mix. nodes. utility.min.node nodes name. Examples utility.max. The function must then return the corresponding aggregated value or utility. name.levels Arguments name. length(par)). ) .node. . name of the function to be used for aggregation.aggregate.fun.aggregation.8).3.aggregate. FALSE.mult(c( .fun name of the node to be constructed as a character string.create Construct an aggregation node Description Function to construct an aggregation node for value or utlity functions. list of nodes to be aggregated.create See Also Constructor of aggregation node: utility. respectively. utility. .2.

Environmental Modelling & Software. P. aggregation at the next higher level is not possible if this node returns NA.min for minimum aggregation utility. Default value is "black". (optional) vector of parameter names corresponging to the vector of values specified under par. Default value is FALSE.aggregate.aggregation.. and Langhans.aggregate.aggregate.mix for mixed aggregation of additive. Textbooks on the use of utility and value functions in decision analysis: .76 utility. (optional) logical variable indicating if the value of this node is required for aggregation at the next higher level. required num. Only required to provide access to the values through a named parameter vector. in press.fun. Constructing.max for maximum aggregation utility. (optional) color used for plotting the bounding box of the node in the objective hierarchy.aggregate. Default value is . evaluating and visualizing value and utility functions for decision support.reichert@eawag. minimum.aggregate. S..mult for multiplicative aggregation utility. number of lower-level values or utilities that must at least be available to make the evaluation possible.aggregate.aggregation with the properties specified in the arguments of the function. and Cobb-Douglas aggregation. (optional) number of hierarchical levels by which the node in the objective hierarchy is shifted to make a branch fit better to other branches. Author(s) Peter Reichert <peter.create utility.required col shift. Schuwirth.levels Value The function returns the created object of type utility.add for additive aggregation utility.ch> References Short description of the package: Reichert. par names.par numeric vector of parameter values to be passed to the function specified unter name. N. If this variable is TRUE.cobbdouglas for Cobb-Douglas aggregation utility.

T.intpol. name.levels = data.create.create. utility. utility = FALSE) # define 1d interpolation end node for bed modification with # riprap # (attribute "bedmodfract_percent" with levels from to 1 ) bedmod_riprap <utility. Berlin.endnode. L. Decisions with Multiple Objectives.. H. R. utility. or utility.intpol1d."moderate". F.utility. Examples # define discrete end node for width variability # (attribute "widthvariability_class" with levels "high"."none")).utility."u.endnode.attrib = "bedmodfract_percent".cond. # "moderate" and "none") widthvar <utility.endnode. and Raiffa. Eisenfuehr. required = FALSE. ). Create end nodes with utility. evaluate. . .intpol1d.endnode. summary.endnode.aggregation.node = "width variability". First published in 1976.discrete. M.create( name.intpol2d.discrete. Weber.create. names.conversion.4125.aggregation.create 77 Keeney.node = "bed modification riprap".aggregation.high"."u_moderate"..aggregation. See Also Print. Cambridge University Press. attrib. or utility. 1993.endnode.create.none"). Springer.utility.create.parfun.create. Create conversion nodes with utility.u = c("u. Rational Decision Making.frame(widthvariability_class= c("high".parfun1d.utility.create.conversion.aggregation and plot. and Langer.create( name. 2010. evaluate and plot the node with print.endnode.utility. u = c(1. utility.

endnode.intpol1d. u = c(1. utility = FALSE) # define 1d interpolation end node for bank modification with # permeable material # (attribute "bankmodfract_percent" with levels from to 1 ) bankmod_perm <utility.6 .frame(bedmodtype_class= c("riprap".1 . .node = "bed modification".1 ). . required = FALSE.1 ). u = c(1.endnode.675. x = c( .aggregation. x = c( . name.1 ). .5625.1 ). utility = FALSE) # define combination end node for bed modification # (attributes "bedmodtype_class" and "bedmodfract_percent") bedmod <utility.cond.bedmod_other). range = c( .775."other")).node = "bed modification other".5625.endnode. . . FALSE) utility.intpol1d.775. nodes = list(bedmod_riprap.78 range x u required utility = = = = = c( .1 . . FALSE.intpol1d.3 . c( . .levels = data. .create # define 1d interpolation end node for bed modification with # other material # (attribute "bedmodfract_percent" with levels from to 1 ) bedmod_other <utility. range = c( .4125.24).attrib = "bankmodfract_percent".8667.24).1 ).create( name.3 . required = FALSE.create( name. utility = FALSE) # define 1d interpolation end node for bank modification with # impermeable material # (attribute "bankmodfract_percent" with levels from to 1 ) bankmod_imperm <utility. required = FALSE.create( name. . attrib.create( name.1 ).attrib = "bedmodfract_percent". .node = "bank modification perm".1 . ). c(1.endnode. name.node = "bank modification imperm".3 .

create name.16).frame(bankmodtype_class= c("perm". nodes = list(bankmod_perm. . .5625. c(1. c( . list(x=c( . )."artificial")).3 ))).node = "riparian zone veg. utility = FALSE) # define discrete end node for riparian zone vegetation # (attriute "riparianzoneveg_class" with levels "natural". .775. list(x=c( . )."riparianzonewidth_m"). ).1.levels = data.aggregation. utility = FALSE) # define aggregation node for riparian zone riparzone <utility.discrete. utility = FALSE) # define 2d interpolation end node for riparian zone width # (attributes "riparianzonewidth_m" and "riparianzonewidth_m") riparzone_width <utility. u = c( . # "seminatural" and "artificial") riparzone_veg <utility.y=c(5.create( name.create( name. FALSE) 79 # define combination end node for bank modification # (attributes "bankmodtype_class" and "bankmodfract_percent") bankmod <utility.utility.6 . attrib.attrib = c("riverbedwidth_m".15.15)). ranges = list(c( .levels = data. c( .1.create( name.y=c(15. u = c(1.".endnode.15)).attrib range x u required utility = = = = = = "bankmodfract_percent".intpol2d.5625.1 ). .6.y=c(3 ."seminatural".5.3 )). .endnode. isolines = list(list(x=c( .2.node = "riparian zone width".endnode. required = FALSE.frame(riparianzoneveg_class= c("natural". FALSE.16).cond. )). lead = 1.16). .1 ).16).bankmod_imperm). attrib.create( name. required = FALSE.3 .node = "bank modification".1 .16).node = "riparian zone". list(x=c( ."imperm")).aggregation.y=c( . name.1 . .24.c( . .

"w_min".create( name. "riprap". 5 .25.aggregation. "bed modification other". name. .mix".1).frame(widthvariability_class bedmodtype_class bedmodfract_percent bankmodtype_class bankmodfract_percent riverbedwidth_m = = = = = = "none".bankmod.par=c(u_moderate= . required = TRUE) # print individual definitions print(widthvar) print(bedmod) # print all definitions print(morphol) # plot objectives hierarchy with attributes plot(morphol) # plot individual nodes: plot(widthvar) plot(widthvar.data. .aggregation. names.riparzone_veg). .25.fun = "utility. par = c(1. nodes = list(widthvar.25.bedmod."w_bedmod"."w_cobbdouglas"). .aggregate.nodes=c("width variability". name.80 nodes = list(riparzone_width. .cobbdouglas"."w_bankmod". 1 .node = "ecomorphology". "bank modification")) # evaluate value function for data sets and plot colored hierarchies # and table attrib_channelized <. 7 . .riparzone). "w_add".25.par = c("w_widthvar".aggregate.2)) plot(bedmod_other) plot(bankmod) #plot(riparzone_width) # plot selected node definitions of a hierarchy plot(morphol.type="nodes".fun = "utility. "imperm". par = c( .create morphol <utility.1)."w_riparzone". required = FALSE) # define aggregation node for ecomorphological state utility.

riparianzoneveg_class = c("seminatural". 15.u=res_rehab."seminatural")[rbinom(sampsize. riparianzoneveg_class = c("natural".mean=1 .sd=5).sampsize). bankmodfract_percent = rnorm(sampsize.u=res_channelized) plot(morphol.sd=5).evaluate(morphol. "seminatural") "high". riverbedwidth_m = rnorm(sampsize.create riparianzonewidth_m riparianzoneveg_class <. "imperm".attrib=attrib_rehab_unc) plot(morphol. "riprap". par=c(w_add=1.sd=15).u=res_channelized_unc) .data. 2 .attrib=attrib_channelized) res_channelized_add <.uref=res_channelized) plot(morphol. . .aggregation.u=res_both.type="table") # consideration of uncertain attribute levels (higher uncertainty for # predicted state after rehabilitation than for observed channelized state): sampsize <.5)+1]) attrib_rehab_unc <.frame( widthvariability_class = rep("high".attrib=attrib_channelized_unc) res_rehab_unc <. bankmodtype_class = rep("imperm".1. riparianzonewidth_m = rnorm(sampsize. riverbedwidth_m = rep(1 . bankmodfract_percent = rnorm(sampsize.evaluate(morphol.sd=5).sampsize).c("channelized".evaluate(morphol.5)+1]) res_channelized_unc <."artificial")[rbinom(sampsize. bedmodtype_class = rep("riprap".2 attrib_channelized_unc <."rehabilitated") plot(morphol. bedmodfract_percent = rnorm(sampsize. "natural") 81 attrib_rehab res_channelized <.sampsize).w_cobbdouglas= )) res_rehab <. bedmodfract_percent = rnorm(sampsize.sampsize). riparianzonewidth_m = rep(5.frame(widthvariability_class bedmodtype_class bedmodfract_percent bankmodtype_class bankmodfract_percent riverbedwidth_m riparianzonewidth_m riparianzoneveg_class = = = = = = = = = = 5.res_rehab) rownames(res_both) <. bankmodtype_class = rep("imperm".u=res_channelized_add) plot(morphol."high")[rbinom(sampsize.mean=5 .mean=7 .attrib=attrib_rehab) res_both <.1.data.1.frame( widthvariability_class = c("moderate".data.sd=2).sampsize).attrib=attrib_channelized.sampsize).w_min= .evaluate(morphol. . 5 .sampsize).mean=5 .rbind(res_channelized. 15.utility.5)+1].evaluate(morphol.mean=2 . bedmodtype_class = rep("riprap".u=res_rehab) plot(morphol.mean=1 .sd=2).

82
plot(morphol,u=res_rehab_unc) plot(morphol,u=res_rehab_unc,uref=res_channelized_unc)

utility.calc.colors

utility.calc.colors

Color Scheme for Value Functions

Description Function to calculate a color scheme for value functions. Usage utility.calc.colors(n = 5) Arguments n Details For n = 5 this function produces the standard colors red, orange, yellow, green and blue as used in river assessment programs. These colors are provided in a lighter version to improve readability of black text in front of the colored background. For large values of n quasi-continuous transitions are defined between these colors. Any other vector of colors can be used by the plotting routines. Value Character vector of colors. Author(s) Peter Reichert <peter.reichert@eawag.ch> References Short description of the package: Reichert, P., Schuwirth, N. and Langhans, S., Constructing, evaluating and visualizing value and utility functions for decision support, Environmental Modelling & Software, in press. Textbooks on the use of utility and value functions in decision analysis: Keeney, R. L. and Raiffa, H. Decisions with Multiple Objectives. Cambridge University Press, 1993. First published in 1976. Eisenfuehr, F., Weber, M. and Langer, T., Rational Decision Making, Springer, Berlin, 2010. number of colors.

utility.conversion.intpol.create See Also See plot.utility.endnode.discrete plot.utility.endnode.intpol1d plot.utility.endnode.parfun1d plot.utility.endnode.intpol2d plot.utility.endnode.cond plot.utility.aggregation plot.utility.conversion.intpol plot.utility.conversion.parfun for the use of such color vectors in plotting functions and utility.get.colors for getting colors corresponding to specified values. Examples
utility.calc.colors(5) utility.calc.colors(1 )

83

utility.conversion.intpol.create Construct an interpolation conversion node

Description Function to construct a node converting values into utilities by interpolation. Usage utility.conversion.intpol.create(name.node, node, x, u, names.x names.u required col shift.levels Arguments name.node node x u name of the node to be constructed as a character string. value node that is to be converted into a utility node. numeric vector of values for which the utility is known. numeric vector of utilities corresponding to the values given in the previous argument x.

= = = = =

rep(NA, length(x)), rep(NA, length(u)), FALSE, "black", )

84 names.x

utility.conversion.intpol.create (optional) vector of character strings with names of the components of the numeric vector x specified above. Only required to provide access to the values through a named parameter vector. (optional) vector of character strings with names of the components of the numeric vector u specified above. Only required to provide acces through a named parameter vector. (optional) logical variable indicating if the value of this node is required for aggregation at the next higher level. If this variable is TRUE, aggregation at the next higher level is not possible if this node returns NA. Default value is FALSE. (optional) color used for plotting the bounding box of the node in the objective hierarchy. Default value is "black". (optional) number of hierarchical levels by which the node in the objective hierarchy is shifted to make a branch fit better to other branches. Default value is .

names.u

required

col shift.levels

Value The function returns the created object of type utility.conversion.intpol1 with the properties specified in the arguments of the function. Author(s) Peter Reichert <peter.reichert@eawag.ch> References Short description of the package: Reichert, P., Schuwirth, N. and Langhans, S., Constructing, evaluating and visualizing value and utility functions for decision support, Environmental Modelling & Software, in press. Textbooks on the use of utility and value functions in decision analysis: Keeney, R. L. and Raiffa, H. Decisions with Multiple Objectives. Cambridge University Press, 1993. First published in 1976. Eisenfuehr, F., Weber, M. and Langer, T., Rational Decision Making, Springer, Berlin, 2010. See Also Print, evaluate and plot the node with print.utility.conversion.intpol, summary.utility.conversion.intpol, evaluate.utility.conversion.intpol and plot.utility.conversion.intpol. Create other conversion nodes with utility.conversion.parfun.create. Create end nodes with

Usage utility. The parametric function must have the arguments u and par which pass a vector of values and a vector of parameters to the function. name.create. utility.parfun.conversion. If this variable is TRUE.endnode. = = = = rep(NA. respectively. Create aggregation nodes with utility.fun.fun.conversion.create. FALSE.levels Arguments name. node.endnode.node node name.create. name of the parametric function to be evaluated as a character string. The function has to return a vector of corresponding utilities.discrete.cond.create. or utility. numeric vector of parameter values to be passed to the function specified unter name. Only required to provide access to the values through a named parameter vector. utility.create Construct a parametric function conversion node Description Function to construct a node converting values into utilities by a parametric function.endnode.parfun. names.create(name.intpol2d.parfun1d. utility. (optional) vector of parameter names corresponging to the vector of values specified under par.par required . value node that is to be converted into a utility node. par. aggregation at the next higher level is not possible if this node returns NA. utility.create. Default value is FALSE.parfun1d.parfun.conversion.par required col shift.endnode.endnode.create 85 utility.utility. length(par)).fun name of the node to be constructed as a character string. (optional) logical variable indicating if the value of this node is required for aggregation at the next higher level.node.aggregation. ) par names. "black".create.

Value The function returns the created object of type utility.parfun. M.conversion.parfun with the properties specified in the arguments of the function.86 col shift.intpol2d.conversion.conversion. in press. F. utility. Springer. N.discrete.utility. Weber.. S.parfun1d. T. and Langhans. 1993. utility. Berlin.create.conversion. Default value is "black".reichert@eawag.intpol. Default value is .conversion. or utility. 2010.utility. (optional) number of hierarchical levels by which the node in the objective hierarchy is shifted to make a branch fit better to other branches.conversion. and Langer.aggregation.create. Create end nodes with utility. L.cond.endnode.parfun and plot. Schuwirth.utility.create.create (optional) color used for plotting the bounding box of the node in the objective hierarchy.endnode. and Raiffa. utility. Constructing. Cambridge University Press. evaluating and visualizing value and utility functions for decision support. evaluate and plot the node with print. Author(s) Peter Reichert <peter. summary.parfun.ch> References Short description of the package: Reichert. First published in 1976.endnode.create. Environmental Modelling & Software. See Also Print. evaluate.endnode.. R.utility. Eisenfuehr..parfun1d.create. Create aggregation nodes with utility. Textbooks on the use of utility and value functions in decision analysis: Keeney.create.conversion..levels utility.create.endnode.parfun. H. .parfun. Create other conversion nodes with utility. Rational Decision Making. Decisions with Multiple Objectives. P.

) nodes utility required col shift.node attrib.create Construct a conditional end node Description Function to construct a node that makes a choice between given end nodes based on the levels of discrete attributes. As combinatorics can lead to a very large number of possible combinations. nodes.cond. Default value is . If this variable is TRUE. (optional) logical variable indicating if a value function (FALSE) or a utility function (TRUE) is created.levels.levels name of the node to be constructed as a character string. Default value is "black".cond.levels above containing the nodes to be associated with the attriute level combinations specified in the rows of attrib. TRUE. in particular if each attribute has many different levels expressed by numbers or character strings. "black".endnode. (optional) logical variable indicating if the value of this node is required for aggregation at the next higher level.utility. Default value is FALSE.levels Value The function returns the created object of type utility. This may be a dependence on any number of attributes.cond with the properties specified in the arguments of the function. vector of the length of the number of columns of the data frame specifed as argument attrib. utility = required = col = shift.create 87 utility. the node should not depend on a too large number of attributes.levels. Usage utility. (optional) number of hierarchical levels by which the node in the objective hierarchy is shifted to make a branch fit better to other branches. aggregation at the next higher level is not possible if this node returns NA.node.create(name. (optional) color used for plotting the bounding box of the node in the objective hierarchy. data frame with attribute names as column names and all discrete attribute level combinations in the rows.cond.endnode.endnode. FALSE. Default value is TRUE. attrib.levels = Arguments name. .endnode.

create. Examples bedmod_riprap <utility.intpol1d.parfun. Berlin. Springer.conversion. H.create. . .cond and plot. utility. See Also Print.5625. utility. R. 1993.aggregation. Create other end nodes with utility.1 .utility.create. First published in 1976.utility.attrib = "bedmodfract_percent". L. x = c( . and Langer. or utility. evaluate..utility.intpol1d. P.utility. evaluating and visualizing value and utility functions for decision support. T. Eisenfuehr.create Reichert.intpol2d. Schuwirth. M.endnode. evaluate and plot the node with print.cond. Create other types of nodes with utility. summary. Weber. utility.endnode.discrete. S..create.cond.endnode.endnode. .create( name.1 ).endnode. Environmental Modelling & Software.cond.intpol. Rational Decision Making.88 Author(s) Peter Reichert <peter.endnode.create.. range = c( .reichert@eawag. 2010.775. or utility.1 ).endnode. N.create.. u = c(1.ch> References Short description of the package: utility.cond. Cambridge University Press. in press.conversion.24). and Langhans. required = FALSE.endnode. and Raiffa. Constructing. .3 . name. Decisions with Multiple Objectives.endnode.endnode.parfun1d.create.node = "bed modification riprap". Textbooks on the use of utility and value functions in decision analysis: Keeney. F.

required = FALSE. length(u)). range = c( . attrib."other")). nodes = list(bedmod_riprap.levels. "black".create(name.attrib = "bedmodfract_percent". attrib.utility.node name of the node to be constructed as a character string.3 . u.node = "bed modification".levels = data. TRUE. name.create( name.u = utility = required = col = shift.775.frame(bedmodtype_class= c("riprap".discrete.cond. . FALSE.discrete.levels = Arguments name. ) .endnode. u = c(1. ).node. names. required = FALSE.bedmod_other). utility = FALSE) bedmod <utility.node = "bed modification other". utility = FALSE) print(bedmod) plot(bedmod) utility.endnode.intpol1d.endnode.endnode.5625. Usage utility.1 . rep(NA.create Construct a discrete value or utility end node Description Function to construct a discrete value or utility end node. .1 ).1 ).endnode. x = c( .create( name.create utility = FALSE) 89 bedmod_other <utility.discrete.

discrete. P. 1993. and Langhans. Berlin. R. and Langer. Rational Decision Making.levels Value The function returns the created object of type utility. u names. Default value is "black". (optional) vector of character strings with names of the components of the numeric vector u specified above. S. numeric vector of the length of the number of columns of the data frame specifed as argument attrib. Weber. Default value is TRUE. and Raiffa. F. Only required to provide access to the values through a named parameter vector.. Schuwirth. Environmental Modelling & Software. Springer.levels.u utility required col shift...endnode.levels above specifying the values or utilities corresponding to the rows of attrib. Eisenfuehr. If this variable is TRUE. Author(s) Peter Reichert <peter.levels utility. Constructing. N. (optional) number of hierarchical levels by which the node in the objective hierarchy is shifted to make a branch fit better to other branches. 2010. M. (optional) logical variable indicating if a value function (FALSE) or a utility function (TRUE) is created. in particular if each attribute has many different levels expressed by numbers or character strings.create data frame with attribute names as column names and all discrete attribute level combinations in the rows. L. evaluating and visualizing value and utility functions for decision support. T. (optional) color used for plotting the bounding box of the node in the objective hierarchy. H. Default value is . Default value is FALSE. in press.reichert@eawag. First published in 1976.discrete with the properties specified in the arguments of the function. the node should not depend on a too large number of attributes.90 attrib. (optional) logical variable indicating if the value of this node is required for aggregation at the next higher level. . Cambridge University Press. Textbooks on the use of utility and value functions in decision analysis: Keeney. This may be a dependence on any number of attributes.. aggregation at the next higher level is not possible if this node returns NA. As combinatorics can lead to a very large number of possible combinations.ch> References Short description of the package: Reichert. Decisions with Multiple Objectives.endnode.

create.utility.discrete.discrete.levels = data.create.cond. names.utility.create.endnode. required = FALSE."none")).endnode.discrete and plot."moderate".create.endnode. . utility.high".discrete. summary.utility.endnode.endnode.create( name.endnode.endnode. Create other types of nodes with utility. evaluate.intpol. utility.discrete. utility = FALSE) print(widthvar) plot(widthvar) utility."u.none").parfun1d. attrib.frame(widthvariability_class= c("high"."u_moderate". ).node = "width variability".utility. 91 Examples widthvar <utility.endnode.intpol2d.4125.create.utility.create.create Construct a single-attribute interpolation end node Description Function to construct a single-attribute interpolation end node.parfun. utility.intpol1d.endnode.conversion.intpol1d.intpol1d. u = c(1. or utility.conversion.endnode.aggregation. or utility.endnode. evaluate and plot the node with print.create.create See Also Print. Create other end nodes with utility. .u = c("u.

x names. range. numeric vector with two components specifying the minimum and the maximum of the attribute range. (optional) logical variable indicating if the value of this node is required for aggregation at the next higher level. rep(NA.node name. length(u)).levels Arguments name.u utility required col shift. ) name of the node to be constructed as a character string. Only required to provide acces through a named parameter vector. If this variable is TRUE.endnode. Default value is "black". Default value is FALSE. names. name. u. (optional) number of hierarchical levels by which the node in the objective hierarchy is shifted to make a branch fit better to other branches. names. (optional) color used for plotting the bounding box of the node in the objective hierarchy.u utility required col shift. FALSE. length(x)). (optional) logical variable indicating if a value function (FALSE) or a utility function (TRUE) is created.create(name. Only required to provide access to the values through a named parameter vector.x utility.92 Usage utility. numeric vector of values or utilities corresponding to the attribute values given in the previous argument x. Default value is TRUE.endnode.endnode.intpol1d.attrib. numeric vector of attribute values for which the value or utility is known. TRUE. (optional) vector of character strings with names of the components of the numeric vector u specified above. aggregation at the next higher level is not possible if this node returns NA.levels Value The function returns the created object of type utility. "black".create = = = = = = rep(NA. Default value is .attrib range x u names.intpol1d with the properties specified in the arguments of the function.intpol1d. x. .node. (optional) vector of character strings with names of the components of the numeric vector x specified above. name of the attribute on which the value or utility function depends as a character string.

create.reichert@eawag.endnode.endnode..utility. T. Cambridge University Press. Create other types of nodes with utility.775. or utility. . and Langer. N. and Raiffa. Textbooks on the use of utility and value functions in decision analysis: Keeney.endnode.parfun1d.cond.1 ).intpol1d and plot. Schuwirth. summary. . utility.aggregation.endnode. S.utility. utility. Eisenfuehr.parfun. L.ch> References Short description of the package: 93 Reichert. H. Constructing.conversion. required = FALSE. Decisions with Multiple Objectives. ).create. 1993.1 . Weber.intpol.create.endnode. Create other end nodes with utility.intpol2d.create Author(s) Peter Reichert <peter. evaluating and visualizing value and utility functions for decision support.3 . Examples bedmod_other <utility. R.1 ). F.utility.intpol1d. name.create.endnode. Springer. Berlin.endnode.5625.endnode.discrete..create. evaluate and plot the node with print.endnode. . in press. Rational Decision Making.. x = c( . and Langhans.node = "bed modification other".create.endnode. First published in 1976.create( name. P.intpol1d.attrib = "bedmodfract_percent". See Also Print.utility. evaluate.intpol1d. u = c(1. range = c( . M.create.. or utility.utility.conversion.intpol1d. utility. 2010.intpol1d. Environmental Modelling & Software.

node name. (optional) vector of character strings with names of the components of the numeric vector u specified above.u lead . 2 along lines with constant values of the second attribute. list of two numeric vectors with two components each specifying the minimum and the maximum of the range of the corresponding attribute.create(name. . isolines.create Construct a two-attribute interpolation end node Description Function to construct a two-attribute interpolation end node. numeric vector of the same length as the outer list of the argument isolines specifying the corresponding values or utilities. TRUE.intpol2d.levels Arguments name.(second attribute) values to characterize the shape of the isoline. numeric value specifying which variable is the lead variable for interpolation. "black".intpol2d.create print(bedmod_other) plot(bedmod_other) utility. Usage utility.endnode.(=first attribute) and y. 1 indicates linear interpolation between isolines along lines with constant value of the first attribute.attrib.node. names. = = = = = = rep(NA.attrib ranges isolines name of the node to be constructed as a character string. length(u)).endnode. ranges.intpol2d. u. ) u names. FALSE. Each definition consists of a list with elements x and y that each represents a numeric vector of x.94 utility = FALSE) utility. names of the attributes on which the value or utility function depends as a vector of two character strings. and zero indicates to take the average of these two interpolation schemes.u lead utility required col shift. Only required to provide acces through a named parameter vector. list of isoline definitions. name.endnode.

Cambridge University Press. aggregation at the next higher level is not possible if this node returns NA.create utility required 95 (optional) logical variable indicating if a value function (FALSE) or a utility function (TRUE) is created. Decisions with Multiple Objectives. Default value is FALSE. See Also Print. summary. evaluate and plot the node with print.endnode..create.utility.intpol2d. F. S.intpol2d with the properties specified in the arguments of the function. First published in 1976. If this variable is TRUE. Eisenfuehr.intpol1d.intpol2d. in press..endnode.utility. Berlin.create. Schuwirth. Weber.parfun1d. (optional) logical variable indicating if the value of this node is required for aggregation at the next higher level.utility. N. utility. H. Author(s) Peter Reichert <peter. 2010.utility.intpol2d.endnode. and Langhans. and Raiffa. utility. col shift. Environmental Modelling & Software.endnode... Default value is "black". Default value is .reichert@eawag.levels Value The function returns the created object of type utility. R. Constructing. P. L. Create other end nodes with utility.endnode. evaluate.endnode. T. Springer.intpol2d. (optional) color used for plotting the bounding box of the node in the objective hierarchy. Default value is TRUE.endnode.ch> References Short description of the package: Reichert.create. (optional) number of hierarchical levels by which the node in the objective hierarchy is shifted to make a branch fit better to other branches. and Langer.discrete. 1993. Textbooks on the use of utility and value functions in decision analysis: Keeney.utility.endnode. Rational Decision Making. M. or . evaluating and visualizing value and utility functions for decision support.intpol2d and plot.endnode.

create.intpol2d. names.aggregation. )).endnode.create Construct a single-attribute parametric function end node Description Function to construct a single-attribute parametric function end node.create. list(x=c( .2.15.endnode.parfun1d.create(name.endnode.16).intpol.endnode.16). length(par)).16). . "black".cond. list(x=c( . TRUE. or utility.16). lead = 1.c( .parfun.5.3 )). Usage utility.node = "riparian zone width". range.15)).conversion.1.96 utility.16). list(x=c( . isolines = list(list(x=c( ."riparianzonewidth_m").3 ))). name. ).1 . par. utility.node.create.y=c(15.1.attrib = c("riverbedwidth_m". ) . utility = FALSE) print(riparzone_width) plot(riparzone_width) utility.attrib.create( name.create riparzone_width <utility. FALSE. .6.levels = = = = = rep(NA.parfun1d.conversion.y=c( . ranges = list(c( .15)).create. name.parfun1d.par utility required col shift.endnode.y=c(5. name. Create other types of nodes with utility. Examples utility. .fun.y=c(3 . u = c( .

. Default value is .attrib range name. H. Environmental Modelling & Software. P.par utility required col shift. Default value is FALSE. (optional) logical variable indicating if the value of this node is required for aggregation at the next higher level. (optional) color used for plotting the bounding box of the node in the objective hierarchy.parfun1d. (optional) number of hierarchical levels by which the node in the objective hierarchy is shifted to make a branch fit better to other branches.utility. R. First published in 1976.fun name of the node to be constructed as a character string. numeric vector with two components specifying the minimum and the maximum of the attribute range.endnode. 97 name of the attribute on which the value or utility function depends as a character string. Decisions with Multiple Objectives.endnode.fun. and Raiffa. (optional) vector of parameter names corresponging to the vector of values specified under par. Default value is "black". S.levels Value The function returns the created object of type utility. respectively. (optional) logical variable indicating if a value function (FALSE) or a utility function (TRUE) is created.. Author(s) Peter Reichert <peter. Only required to provide access to the values through a named parameter vector. Default value is TRUE. N.node name. Constructing. name of the parametric function to be evaluated as a character string. Cambridge University Press. Textbooks on the use of utility and value functions in decision analysis: Keeney. The function has to return a vector of corresponding values or utilities. aggregation at the next higher level is not possible if this node returns NA. L. evaluating and visualizing value and utility functions for decision support. par names. Schuwirth.ch> References Short description of the package: Reichert.reichert@eawag..parfun1d with the properties specified in the arguments of the function. numeric vector of parameter values to be passed to the function specified unter name. The parametric function must have the arguments attrib and par which pass a vector of attribute levels and a vector of parameters to the function. If this variable is TRUE. in press.create Arguments name. and Langhans. 1993.

utility. utility = FALSE) print(bedmod_other) plot(bedmod_other) utility.parfun. and Langer.endnode.cond.endnode. T.utility. Weber. M. utility.create( name.endnode.98 utility.create.create.fun. range = c( . evaluate and plot the node with print. 2010.attrib = "bedmodfract_percent".intpol.create.endnode.aggregation.exp Exponential function for value or utility functions Description Exponential function for value or utility functions.intpol2d.fun. ). Berlin.parfun1d and plot.conversion.parfun1d.fun = "utility.parfun1d. name. or utility. name.parfun1d.utility.endnode. See Also Print. required = FALSE.endnode.create.utility.endnode. Create other types of nodes with utility. Create other end nodes with utility.exp Eisenfuehr. summary.conversion.exp".parfun1d.discrete. F.create..endnode.endnode. Examples bedmod_other <utility.1 . or utility. Rational Decision Making.create.create. utility. evaluate. utility.intpol1d.. .node = "bed modification other". par = c(-1.fun. Springer.1 ).

F.. Weber. Examples utility. Eisenfuehr. See Also See the node constructors utility. in press. Value Vector of values or utilities corresponding to the attributes passed by argument a Author(s) Peter Reichert <peter. Springer. Cambridge University Press.conversion. R. Textbooks on the use of utility and value functions in decision analysis: Keeney.create in which this function can be used. S. and Raiffa. H.fun..fun. 2010.utility. Vector of parameters: par[1]: absolute risk aversion par[2]: minimum of attribute range (default = 0) par[3]: maximum of attribute range (default = 1) 99 Details The function evaluates the expression (1-exp(-par[1]*(a-par[2])/(par[3]-par[2])))/(1-exp(-par[1])).exp( :1 /1 . and Langhans.reichert@eawag. Decisions with Multiple Objectives. evaluating and visualizing value and utility functions for decision support.endnode. N... Constructing.exp Usage utility. L. First published in 1976.intpol1d. .1)) .intpol. par) Arguments attrib par vector of attribute levels to calculate corresponding value or utility.exp(attrib.fun. Environmental Modelling & Software. Berlin. 1993. and Langer. Schuwirth. P. M.create and utility. Rational Decision Making.par=c(2. T.ch> References Short description of the package: Reichert.

N. Decisions with Multiple Objectives. Environmental Modelling & Software. Schuwirth.5.colors utility. Cambridge University Press. Value Character vector of colors.colors Examples utility.calc. Eisenfuehr..get.get. 1993.100 utility. H.colors()) Arguments u col value level representing the evaluation ot a value function (this value level has to be between zero and unity). and Langhans. Springer. T.. S. F. P. Textbooks on the use of utility and value functions in decision analysis: Keeney. L. color scheme (vector of colors to be used for a division of the interval between zero and unity into equal intervals)).col=utility.. Author(s) Peter Reichert <peter. and Raiffa.colors Get Color Corresponding to Specified Value Levels Description Function to get the colors from a given color scheme at specific value levels. 2010. Berlin. R. evaluating and visualizing value and utility functions for decision support. M.colors(c( . and Langer.get.calc. in press..ch> References Short description of the package: Reichert.colors(u.reichert@eawag.1)) . . Usage utility. See Also See utility. Constructing. First published in 1976.get. Rational Decision Making. Weber.

5. 43. 93 plot.discrete. 84 print.intpol2d.utility. 17.utility. 86 evaluate. 26. 5.endnode. 62. 35.utility. 24–28. 77 summary. 5.aggregation. utility function utility-package. 83. 33. 4.conversion. 64–67 updatepar. 35.endnode. 62.intpol2d. 38. 5.intpol1d. 4. 17.utility.intpol.endnode.utility. 33. 4.Index ∗Topic decision analysis. value function.utility.endnode. 5. 35. 4. 40. 4.utility. 46.aggregation. 40.utility. 3 evaluate.parfun1d. 64–67 updatepar.utility. 12. 13. 16. 4.utility. 32. 60. 4. 4. 37.discrete.discrete. 98 plot. 15. 84 evaluate. 4. 5. 84 summary.utility. 88 summary. 58. 60–62. 20. 84 plot. 47. 77 evaluate. 4. 91 summary. 93 summary.parfun. 22.intpol2d. 83. objectives hierarchy.utility. 13.intpol.utility. 56. 37. 57 updatepar. 5. 24. 18. 13. 50.utility. 54.endnode. 20. 44.parfun. 23.aggregation.endnode. 38. 36. 40. 29. 95 evaluate.conversion. 33.utility.parfun.utility. 86 plot. 38.parfun1d. 19. 88 print.conversion. 93 print.endnode. 83.parfun1d.intpol.conversion. 23. 38.parfun.cond.utility.intpol1d. 61. 83. 4. 91 print.endnode. 83.utility. 45. 83. 39. 37. 33. 4. 4.intpol1d. 22.utility. 29. 29.endnode. 13. 14. 26. 14. 10 evaluate. 83 plot. 4.endnode. 37.intpol2d.endnode. 42. 4. 5. 59. 49.conversion. 16.conversion.endnode. 4.endnode.utility. 22. 5.utility.endnode.parfun1d.utility. 19. 4.aggregation.utility.conversion. 37. 98 updatepar.utility.endnode.utility. 19.utility. 93 evaluate.utility. 98 101 print. 98 summary.endnode. 91 plot. 35. 77 print. 34. 41.utility. 58–60.utility.conversion.utility. 86 print.conversion. 30.endnode. 64–67 . 40.utility. 58. 5. 37. 27. 23. 58. 32. 35. 51. 21.intpol. 16. 40. 59.conversion.cond. 14.utility.intpol1d.cond. 4. 4. 95 print.parfun. 32. 33.intpol. 19. 16. 88 evaluate. 83. 95 plot. 4.utility.endnode.endnode. 17. 31. 5.cond.discrete.aggregation.endnode. 91 evaluate. 4. 20. 95 summary. 88 plot.utility. 48. 55. 53. 22. 34. 4.utility. 77. 86 summary. 30. 39. 52.

85. 22.utility. 32.conversion. 26. 65.discrete. 98 utility. 3 utility. 58–61. 69. 22. 13. 91. 11.create. 88. 34. 88. 71–73.parfun1d. 34.intpol1d. 87. 86. 91. 19. 16. 17. 83.utility. 96 utility. 20. 46. 23. 67 updatepar.intpol2d.102 updatepar. 71. 19. 73.create. 3. 3. 19. 23. 49. 68.max.endnode. 3. 37. 23. 91.endnode. 86. 89. 14. 75. 71–73. 40.endnode. 71–73. 48.mult. 3. 77. 85.create. 22. 53. 65. 34. 86. INDEX 77. 19. 88. 20. 95.colors. 68. 65–67. 11. 76 utility. 13. 14. 36.endnode.aggregate. 13.colors. 17. 14. 77. 14. 58.create.fun. 63. 23. 20. 23. 96. 20. 13. 41. 98 utility. 68. 76 utility. 96. 58. 64.aggregation. 11. 16. 3. 75.calc. 88. 36. 100 utility. 75. 100 . 68. 66. 40. 68. 43. 98. 19. 58. 16.intpol. 84. 76 utility. 19. 85. 26. 11.exp. 85. 91. 66 utility (utility-package). 50. 20. 74. 29. 69. 93. 73. 39.cond. 75. 67.aggregate. 58–62. 77. 65–67 updatepar. 96.aggregate. 17. 98 utility.endnode. 14.add. 91. 16.create.endnode. 66.create. 59.parfun1d. 17. 11. 47. 82. 37. 56. 69. 17.aggregate. 95. 86. 17. 37.create. 42. 14.aggregate. 51. 58–63. 83. 98 utility. 99 utility. 58–63. 45. 32. 86. 16. 19.create. 32. 39. 40. 20. 88. 58. 35. 68.get. 93.endnode. 96. 17. 58–62. 77. 13. 39. 99 utility. 58. 13. 93. 98. 91. 68. 71.utility. 22. 75. 4. 69.endnode. 77. 20. 77. 93.endnode. 98 utility. 85. 60–63. 14. 95. 88. 65. 75. 3 utility-package.cobbdouglas. 72. 91. 69. 76 utility.utility. 93. 93. 94. 55. 54. 58. 4. 71. 22. 76 utility.intpol2d. 72.conversion. 75. 93.cond. 61. 58. 29. 13. 13. 70. 22. 76 utility. 71–73. 52. 69.mix. 72. 26.endnode. 64–67 updatepar. 14. 23. 29.aggregate. 11.min. 44. 4. 88.discrete.parfun. 64. 86. 16. 11. 16. 11. 23. 22. 58. 67 updatepar.utility. 91. 98 utility. 85.intpol1d.