UCD Smurfit · Brugha Nomological Method · AY 2025-26 · User Manual v2
1. What is DISC-MCDM?
DISC-MCDM is a browser-based decision support tool built on the Nomological MCDM framework developed by Prof. Cathal Brugha at University College Dublin. It provides a structured, transparent and academically rigorous method for making complex decisions involving multiple alternatives evaluated across multiple criteria organised into three Nomological tiers.
The three tiers reflect the three dimensions of human decision-making: Somatic (Having - tangible, measurable criteria), Psychic (Doing - relational and contextual criteria) and Pneumatic (Being - values-based and ethical criteria). The tool supports two scoring methods: DISCUS for independent utility scoring across three or more alternatives, and DISCRIM for fine discrimination between two or three closely-ranked alternatives.
2. Getting Started
Click Start New Analysis on the home screen to begin. You will be taken through a four-step workflow: Setup, Weights, Score, and Results. Each step must be completed in order before proceeding to the next.
If you want to explore the tool first without building your own decision, use Quick-Start Templates at the bottom of Step 1. Select any template and click Load. The template populates the tool with a complete set of alternatives, criteria, weights and sample scores that you can modify freely.
To resume a saved session, click Load in the top bar and select your previously saved JSON file. The tool restores your decision exactly as you left it, including all criteria, weights, scores and the current step.
3. Step 1 - Setup: Alternatives and Criteria
Alternatives are the options under evaluation (e.g. Supplier Alpha, Supplier Beta). Add between 2 and 15 alternatives using the text box and Add button. Click any name to edit it inline. You can reorder alternatives by dragging.
Criteria are the factors you will score each alternative against. Each criterion belongs to one of the three Nomological tiers. To add a criterion, select its tier, type a name and click Add. To add a sub-criterion (creating a two-level hierarchy), click the arrow beside an existing criterion and add a child.
Groups of sub-criteria are collapsed and expanded using the arrow toggle. The criteria tree on the right shows the full hierarchy at all times.
Scoring basis: Set the maximum score (default 10). This defines the scale for all scoring cells. Change it before entering any scores as changing it afterwards clears all existing scores.
4. Step 2 - Weights: Tiers and Criteria
Tier weights control the relative importance of Somatic, Psychic and Pneumatic. Move the sliders until they sum to 1.00. Use Equal to distribute evenly, or Normalise to rescale the current values proportionally to sum to 1.00.
Criterion weights control the relative importance of criteria within each group. Each group of siblings at the same level must independently sum to 1.00. The effective weight shown below each slider is the normalised contribution of that criterion to the overall score - this is what actually matters for the ranking.
The Criterion Contribution donut chart on the right shows the global weight of each tier as a coloured slice. Click any slice to drill into that tier and see how weight is distributed across its criteria. Click a leaf criterion to see per-alternative scores. Click Back to overview to return.
Tip: the global weight of a criterion = tier weight x all ancestor local weights x its own local weight. All global weights across all leaf criteria sum to 1.00.
5. Step 3 - Score: DISCUS Mode
In DISCUS mode (light blue background), score each alternative independently on each criterion from 0 to the basis. A score of 0 means the worst possible performance on that criterion; the basis means the best possible. Scores are completely independent - giving Alt A a 9 does not constrain what you give Alt B.
Criteria are displayed as a tree grouped by tier. Leaf criteria (those with a filled circle icon) are the ones you score directly. Parent group criteria (hollow square icon) are structural groupings that are not scored directly.
The Fill Sample Scores button (navy, top toolbar) populates all cells with random values for exploration. Clear All Scores (red outline) resets all cells to 0. Use the horizontal scrollbar to see all alternatives when there are many.
The amber-highlighted cells show sample scores pre-loaded from a template. They are fully editable - just type your own value.
6. Step 3 - Score: DISCRIM Mode
In DISCRIM mode (dark teal background), for each criterion row you distribute a fixed number of points (the basis, e.g. 10) across all alternatives, reflecting their relative intensity on that criterion. If Alternative A is approximately twice as good as Alternative B, give A roughly 6.7 and B roughly 3.3 - these sum to 10.
The Row Sum indicator on the right end of each criterion row shows the running total. It turns green when the row sums to exactly the basis, and red when it does not. All rows must be green before results can be computed.
The per-cell hint below each input shows the acceptable range: how much remains for that cell given what the others contain. The last alternative in each row shows the exact required value as = X.X.
Tip: switch between DISCUS and DISCRIM using the Switch button on the Results page. Scores for each method are stored independently - switching does not erase your other-method scores.
7. Step 4 - Results: Rankings and Visualisations
After clicking Compute Results, the Results page shows: a ranked list of alternatives with proportional bars; a Tier Contribution chart breaking the score into Somatic, Psychic and Pneumatic components; advanced visualisations (Advantage Heatmap and Radar Chart); a Score Breakdown Tree; and a Sensitivity Analysis panel.
The Advantage Heatmap compares the top-ranked alternative against each opponent criterion by criterion. Blue cells mean the winner is ahead; amber cells mean the winner is behind. Darker colour means a larger gap. Use this to identify the winner weakest criteria - those could be deal-breakers.
The Radar Chart shows each alternative as a polygon across all criteria axes. Toggle alternatives on and off using the coloured buttons. Overlapping polygons reveal where alternatives are similar or diverge.
8. Score Breakdown Tree
The Original Weighted Score Breakdown is a horizontal tree diagram flowing left to right: Decision node, then tier nodes, then criteria nodes, with score cells attached to each leaf criterion. Each leaf shows a weight badge (global weight as a percentage) and one score cell per alternative containing the raw score and the weighted contribution (raw x global weight).
Scroll horizontally to see all alternatives. Node colours become progressively lighter with increasing depth so you can see the hierarchy level at a glance. The Sensitivity Weighted Score Breakdown below it shows the same structure recalculated using the current sandbox slider weights.
9. Sensitivity Analysis
The Sensitivity Analysis panel lets you explore how robust the ranking is to changes in criterion weights without altering your saved decision. Move any slider to increase or decrease that criterion weight in a sandbox copy. The rankings, charts and breakdown update live.
The Breakeven value shown below each slider is the weight at which the current leader would be overtaken. A value of 0.47 means: if this weight rises above 0.47, the ranking flips. A dash means no flip occurs across the full range - the ranking is stable for this criterion.
The Tornado Chart ranks criteria by influence: the longer the bar, the more the ranking is affected when that criterion weight is swept from 0 to 1. Focus attention on the top two or three bars when assessing robustness.
Click Reset Weights to restore your original saved weights at any time. Your actual saved weights are never altered by the sliders.
10. Magnifying Glass
When two alternatives score very close overall (less than 5% apart), the Magnifying Glass (O Brien and Brugha 2010) automatically appears. It allows you to focus the evaluation on one sub-branch of the criteria tree, renormalising weights within that branch to reveal subtle differences the full model cannot discriminate.
Select a top-level criterion branch from the dropdown and click Focus This Branch. The rankings update to reflect only that branch with renormalised weights. Click Exit Magnifying Glass to restore full-tree rankings.
11. Refining the Decision (Stage 7 - Pliability)
The Refine Decision panel at the bottom of the Results page implements Stage 7 (Pliability) of the Brugha 8-stage DISC cycle. After seeing initial results, you can revise your decision before committing to a final recommendation.
Revise Scores returns to Step 3. Revise Weights returns to Step 2. Modify Alternatives opens a panel where you can remove an existing alternative or add a new one. Removed alternatives are cached and can be restored with their full original scores using the Re-add button.
All refinement actions auto-save a checkpoint first. You can restore any previous state using the Session Version History in the left sidebar.
12. Saving and Loading Sessions
Click Save in the top bar to download your session as a JSON file. On supported browsers (Chrome, Edge) a native Save As dialog appears so you can choose the file location. On other browsers the file downloads automatically to your Downloads folder.
Click Load to restore a previously saved session. The tool keeps a version history of up to 30 automatic checkpoints in your browser (listed in the left sidebar). You can restore any checkpoint by clicking it - this does not affect the saved file on disk.
Use Export CSV (on the Results page) to download a spreadsheet of the final rankings and scores for use in reports or presentations.
13. References
Brugha, C.M. (2004) Phased multicriteria preference finding. European Journal of Operational Research 158, 308-316.
Brugha, C.M. (2004) Structure of multi-criteria decision-making. Journal of the Operational Research Society 55, 1156-1168.
O Brien, F.A. and Brugha, C.M. (2010) The Magnifying Glass method for decision-making. Journal of the Operational Research Society.
Kakeneno, J.R. and Brugha, C.M. (2017) Usability of Nomology-based Methodologies. CEJOR 25, 393-415.
Branigan, C. and Brugha, C. (2013) Behavioural Biases on Residential House Purchase Decisions. UCD Working Paper.
