Table 10 Scenario complexity scoring specifications

From: Bridging urban theory and artificial intelligence: a multi-agent recommendation system for sustainable city development

Dimension

Measurement

Weight (wi)

Normalization

c1: Domain complexity

Unique technical domains

0.20

\(\min (n/5,1)\)

c2: Objective multiplicity

Competing objectives

0.18

\(\min (n/4,1)\)

c3: Constraint severity

Hard constraints count

0.15

\(\min (n/4,1)\)

c4: Stakeholder diversity

Shannon entropy

0.15

\(H/ln(6)\)

c5: Temporal scope

Planning horizon (years)

0.12

\(\min (T/10,1)\)

c6: Spatial boundaries

Categorical scale

0.12

See notea

c7: Data heterogeneity

Distinct data types

0.08

\(\min (n/6,1)\)

  1. aBlock = 0.2, Neighborhood = 0.4, City = 0.6, Region = 0.8, Multi-region = 1.0.
  2. Weights calibrated via Delphi method (12 experts, 3 rounds, Kendall’s W = 0.82).
Back to article page