Home page   Foundation   Our work   Key domains   Project failures   Standards evolution   Data reference models   Recommendations   Leading edge   Online services   Contact

Decision analysis model building

The following DRM represents the classical decision analysis cycle model refinement following the reiterative evaluation of deterministic, probabilistic and informational phases following R. Howard, Decision Analysis Approach, SRI & Stanford University, 1968-2015)

This section is being updated with a range of recent applications for DRMs including:
  • over 300 policy annual report analytical calculations for an annual agicultural report
  • over 80 analytical tools for project design to comply with Agenda 2030 Sustainable Development Goal objectives

Data Reference Models can be used for a wide range of applications. Below there are three representations for:
  • a general schemata of functions that would normally be placed into a DRM
  • a process model such as a manufacturing process
  • a policy model that requires specific information

Experimental and survey design needs to take into account the physical distribution of the objects whose properties need to be measured to generate the appropriate dataset to acquire the desired knowledge on specific relationships. In the case of river basins and analysis of water flux, the existing catchment area is normally divided into segments distributed down the complete run of the river within which stratified data is to be collected.

General logic for river basin data

For each specific aspect of knowledge, if there is a requirement for a different dataset and associated algorithms, it is advisable to create a separate DRM to each output requirement. This helps maintain transparency on the total dataset requirement which is made up of the sum of DRM datasets after deducting duplicate data. In natural resource systems there is normally a variance in data according to the geographic locational and altitude and timing of data collection within a day or over the year. In data collection, periodic frequency can have important implications for the utility of the data in the knowledge generation models for which it is intended.