|
DEXi is a computer program for creating multi-attribute decision models (MADMs) that:
- allows interactive development of qualitative multi-attribute decision models
- allows comparisons of differing attribute options
- can be used to support complex decision-making tasks, i.e. selection of best option from a
range of possible options
- is used to create hierarchical, multi-attribute models from the decomposition of problems
into smaller, less complex sub-problems that may be solved more easily
| MADMs, like DEXi, are well suited to use in assessing crop production systems
that are reliant on complex networks of interactions and trade-offs between biotic,
abiotic and economic components.
DEXi-CSC has been developed to assess systems such as the CSC's integrated cropping
system directly against standard commercial practice over a six year croping sequence.
The effect of cropping treatment was assessed according to the responses of a suite of
indicators which were used to parameterise a qualitative multi-attribute model.
Scenarios were run to test whether the integrated cropping system achieved greater
levels of overall sustainability relative to standard commercial practice.
This case study demonstrates the value of a whole-systems approach using qualitative
multi-attribute modelling for the assessment of existing cropping systems and for
predicting the likely impact of new management interventions on arable sustainability.
|
|
|
DEXi-CSC is comprised of 97 input variables representing the basic biotic, abiotic and
financial indicators of the arable cropping system (see DEXi-CSC input attributes).
Input variables are aggregated into a
hierarchical tree with a total of 332 aggregate variables.
Overall sustainability,
the top level, is the aggregate of two, second level branches
Environmental sustainability and
Economic sustainability.
Environmental sustainability
is the aggregate of three, third level branches
(Biodiversity,
Losses and
Resource use).
Economic sustainability
is the aggregate of two, third level branches
(Viability and
Real profitability).
These branches reflect the main goals of the CSC platform:
- maintainence of yield whilst enhancing biodiversity (in-field and field margin dicot
weeds that benefit invertebrates)
- increasing resource use efficiency (reduce reliance on non-renewables)
- minimise losses via leaching, run-off, erosion and greenhouse gas emissions
|
DEXi-CSC input attributes
- Average market price
- Average price of fuel
- Baling fuel
- Carting fuel
- Connectivity of semi-natural habitats
- Crop competition
- Compost fuel
- Conditions for spray drift
- Conditions for volatilisation
- Cover crop fuel
- Crop cover
- Crop fuel
- Crop rotation diversity/intensity
- Crop type
- Deep inversion tillage fuel
- Desiccants AI/rate
- Desiccant fuel
- Environmental subsidies
- Field margin floral diversity
- Field margin structure
- Field properties
- Fungicide AI/rate
- Fungicide fuel
- FYM fuel
- Harvesting fuel
- Herbicide AI/rate
- Herbicide fuel
- Insecticide AI/rate
- Insecticide fuel
- Inversion tillage fuel
- Irrigation availability
- Irrigation fuel
- Irrigation requirement
|
- Labour hourly wage
- Lime fuel
- Mineral fertilisers fuel
- Molluscicide AI/rate
- Molluscicide fuel
- N soil residual
- Non-environmental subsidies
- Non-inversion tillage fuel
- Number of hours
- P soil residual
- Pathogen pressure
- Pest pressure
- Post-plant herbicide AI/rate
- Precipitation
- Price of compost
- Price of cover crop seed
- Price of crop seed
- Price of desiccants
- Price of fungicide
- Price of FYM
- Price of herbicide
- Price of insecticide
- Price of irrigation
- Price of lime
- Price of mineral K fertiliser
- Price of mineral N fertiliser
- Price of mineral P fertiliser
- Price of molluscicide
- Price of seed treatment
- Price of soil sterilants
- Price of straw
- Price of trace elements
- Price of undersow seed
|
- Product quality
- Production risk
- Proportion of gross margin due to main crop
- Rate compost
- Rate FYM
- Rate lime
- Rate mineral K fertiliser
- Rate mineral N fertiliser
- Rate mineral P fertiliser
- Rate straw
- Rate trace elements
- Rate undersow
- Ratio semi-natural: cultivated land
- Requirement for agricultural equipment
- Seed cover crop
- Seed rate crop
- Seed treatment AI/rate
- Seed undersow
- Soil cover at pesticide application
- Soil sterilants AI/rate
- Soil sterilant fuel
- Soil type
- Straw chopping fuel
- Straw selling price
- Straw yield
- Stubble/straw management
- Tillage intensity
- Trace elements fuel
- Undersow fuel
- Weather conditions
- Weed management strategy
|
The scales used for each input and aggregate function are available
here.
An interactive Java implemntation of the DEXi-CSC model can be accessed here.
Interpreting DEXi Evaluations
Viewing and interpreting several
evaluations of a DEXi tree with over 300
aggregate functions can be challenging. To aid interpretation we use two types of visual
representation of the output i.e. heat maps and radar plots.
DEXi-CSC heat map
These heat maps show the top seven layers of evaluation output for each of the six crops in the CSC
the rotation. Input data is aggregated across six years of the first crop rotation.
As a contrast conventional potato performs poorly whilst Integrated bean performs well.
|
Bean |
Potato |
Spring barley |
Winter barley |
Winter
oilseed rape |
Winter wheat |
| Conventional |
 |
| Integrated |
| |
|
| |
These radar plots show selected attributes from layers three and four of evaluation output for each
of the six crops in the CSC the rotation.
Input data is aggregated across six years of the first crop rotation.
| Bean |
Potato |
Spring barley |
|
|
|
| Winter barley |
Winter oilseed rape |
Winter wheat |
|
|
|
| |
| Integrated |
Overlap |
Conventional |
|
| |
DEXi web site
| |
