Singapore public transport landscape

Yes. In fact, Singapore is a very suitable real-world test bed for turning your framework from a conceptual “quantum sustainability” research agenda into a full-scale, multi-stakeholder network-analysis research programme.

The key is to make the classical network/behavioural model the baseline, and let quantum or quantum-inspired methods compete against it rather than assuming quantum advantage.

Singapore already provides unusually strong conditions for such a study: a highly integrated public-transport network, substantial government transport data, established behavioural research, environmental targets, and a policy objective of increasing sustainable transport adoption. The Green Plan targets 75% peak-period mass-public-transport modal share by 2030 and more than 80% by 2040.

Proposed research model

Singapore Community Public-Transport Adoption Network

The central research question could be:

How can Singapore increase sustained public-transport adoption by understanding the interaction between citizens, behaviour, social networks, transport infrastructure, environmental motivation, government policy and technology — and can quantum-inspired optimisation improve intervention design compared with classical methods?

This gives your original framework a much stronger empirical structure.

---

1. The six-layer network

I would model the ecosystem as six interacting networks, rather than treating “the community” as one homogeneous network.

Layer	Main actors	What we measure
1. Citizens	commuters, families, students, elderly, workers	mode choice, habits, cost, time, comfort, reliability
2. Behaviour science	NUS, universities, behavioural researchers	motivation, social norms, habit formation, intervention response
3. Research institutions	NUS, NTU, research centres	causal models, experiments, network science, optimisation
4. Environmental organisations	NGOs, community groups, sustainability organisations	climate awareness, environmental campaigns, citizen mobilisation
5. Government	MOT, LTA, NEA and related agencies	policy, infrastructure, incentives, regulation, data
6. Technology / operators	transport operators, mobility platforms, AI/data companies	routing, predictive analytics, optimisation, digital engagement

This is important because adoption is not simply a citizen decision.

A commuter's behaviour depends on:

individual preference → household → social norms → neighbourhood → transport accessibility → service reliability → price → policy → information → environmental awareness → technology.

That is a network problem.

---

2. Singapore already gives you a natural empirical foundation

This is not merely hypothetical.

Singapore's LTA currently reports a rail system exceeding 140 stations across six MRT lines, with more than three million daily rail trips, while the broader public transport system also includes buses and LRT.

LTA also maintains public datasets covering annual and monthly public-transport ridership.

More importantly, Singapore already has behavioural research that fits your proposed framework.

For example, NUS/LTA research has examined transport choice using factors including:

• travel needs
• cost
• time
• convenience
• habit
• affect
• information certainty
• stress
• social norms
• environmental sustainability attitudes.

Another NUS study used smart-card data to examine travel-time uncertainty and found that reliability mattered particularly to working adults, while fare cost was more important for seniors, students and children.

So your framework can connect to an existing research base rather than starting from zero.

---

3. The citizen network

This should be the bottom layer of the model.

Instead of simply asking:

"Does this person use public transport?"

measure a vector:

\[
C_i =
[
F_i,T_i,R_i,K_i,H_i,S_i,E_i,A_i
]
\]

where:

• \(F\) = fare sensitivity
• \(T\) = travel-time sensitivity
• \(R\) = reliability sensitivity
• \(K\) = convenience
• \(H\) = habitual behaviour
• \(S\) = social influence
• \(E\) = environmental motivation
• \(A\) = accessibility

Then model the probability of choosing public transport:

\[
P(PT_i)=
f(F_i,T_i,R_i,K_i,H_i,S_i,E_i,A_i)
\]

This is already a conventional behavioural-science model.

That should be your baseline.

---

4. Then introduce the social network

The individual is not isolated.

Suppose:

A → B → C → D

represents social influence.

A commuter may observe:

• colleagues taking MRT;
• friends participating in car-free campaigns;
• family members changing travel habits;
• influencers discussing climate;
• employers providing public-transport incentives.

Then:

\[
Adoption_i(t+1)
=
f(
Adoption_i(t),
NetworkInfluence_i(t),
Infrastructure_i,
Policy_i,
PersonalPreference_i
)
\]

This lets you investigate an important question:

Does public-transport adoption spread through communities like a behavioural contagion?

Not necessarily literally like disease, but mathematically as network diffusion.

---

5. The government network

Government should not be represented simply as "the policymaker".

It becomes another node in the network.

For example:

MOT → LTA → operators → infrastructure → commuters

and:

Green Plan → public transport → citizen behaviour → emissions

LTA explicitly describes its role as managing traffic flow, public transport reliability, active mobility and the transition toward a car-lite city.

The Bus Contracting Model also gives LTA a central role in planning bus services, while operators deliver services against defined standards.

That makes Singapore particularly interesting for network analysis because institutional and operational relationships are already highly structured.

---

6. Environmental organisation network

This is where your sustainability component becomes more interesting.

Environmental organisations could function as information and norm-amplification nodes.

For example:

\[
EnvironmentalMessage
\rightarrow
EnvironmentalAwareness
\rightarrow
SocialNorm
\rightarrow
Behaviour
\rightarrow
PT Adoption
\]

But the research should test whether this actually occurs.

You should not assume that environmental messaging changes behaviour.

Instead:

Experiment

Group A:

Normal transport information.

Group B:

Transport information + environmental benefits.

Group C:

Transport information + environmental benefits + social comparison.

Group D:

Transport information + environmental benefits + reward.

Then measure:

\[
\Delta PT Adoption
\]

This becomes experimentally testable.

---

7. Technology-company network

Technology companies become the real-time optimisation layer.

Potential functions include:

Prediction

Predict:

\[
Demand_{station,time}
\]

Routing

Optimise:

\[
Route =
f(
time,
crowding,
reliability,
energy,
weather
)
\]

Personalisation

Recommend:

"Take MRT + walk today."

rather than:

"Drive."

Dynamic intervention

For example:

"Your normal car journey takes 38 minutes. MRT + walking is estimated at 41 minutes today, but avoids congestion and produces lower transport emissions."

The system then observes whether the recommendation changes behaviour.

---

8. The research institution becomes the independent evaluator

This is crucial.

NUS/NTU/research institutions should not simply become another advocacy node.

They should sit partly outside the intervention system as the evaluator.

Their role:

Classical modelling

• regression
• discrete-choice modelling
• agent-based modelling
• causal inference
• social-network analysis
• reinforcement learning
• optimisation

Experimental validation

• A/B testing
• RCT
• difference-in-differences
• longitudinal studies

Singapore already has examples of this experimental approach. NUS researchers conducted an RCT involving more than 900 commuters to examine responses to peak-hour pricing interventions.

That gives your proposed framework a credible methodological precedent.

---

9. Now your quantum component becomes much more defensible

This is where I would substantially modify your original proposition.

Don't claim:

"Quantum computing will solve public-transport behaviour."

Instead:

Can quantum-inspired or quantum optimisation methods solve selected intervention-allocation problems more efficiently or produce better solutions than established classical optimisation methods?

That is a scientifically testable question.

For example:

Suppose Singapore has:

100 neighbourhoods

and:

20 possible interventions

such as:

• fare incentives
• awareness campaigns
• improved feeder buses
• cycling connections
• reliability improvements
• gamification
• carbon feedback
• employer incentives.

You have a limited budget:

\[
B = \$10m
\]

You want to maximise:

\[
Impact =
\sum_i
Population_i
\times
\Delta PT_i
\times
EmissionReduction_i
\]

subject to:

\[
Cost \leq B
\]

This becomes an intervention-allocation optimisation problem.

That is a much more credible place to investigate quantum optimisation.

---

10. Classical baseline vs quantum model

Your experiment should have three competing models:

Model A — Classical

Agent-based + behavioural + network model.

Model B — Quantum-inspired

Quantum-inspired optimisation / probabilistic formulation running on classical hardware.

Model C — Quantum

Where suitable quantum hardware/algorithms are actually available.

Then compare:

Measure	Classical	Quantum-inspired	Quantum
solution quality	✓	✓	✓
computation time	✓	✓	✓
scalability	✓	✓	✓
energy consumption	✓	✓	✓
intervention cost	✓	✓	✓
behavioural accuracy	✓	✓	✓
robustness	✓	✓	✓

No assumption is made that quantum wins.

If classical performs better, that is itself a legitimate research result.

---

11. The full feedback loop

This is perhaps the strongest visual architecture for your book.

Conceptually:

                 GOVERNMENT
              MOT / LTA / NEA
                    │
                    ▼
              POLICY & FUNDING
                    │
                    ▼
RESEARCH ───── TRANSPORT SYSTEM ───── TECHNOLOGY
  │             │     │                 │
  │             │     │                 │
  ▼             ▼     ▼                 ▼
BEHAVIOUR     MRT    BUS            AI / DATA
SCIENCE       NETWORK NETWORK       OPTIMISATION
  │             │     │                 │
  └─────────────┴─────┴─────────────────┘
                    │
                    ▼
                 CITIZENS
                    │
          ┌─────────┼─────────┐
          ▼         ▼         ▼
       FAMILY    COMMUNITY   WORKPLACE
          │         │         │
          └─────────┼─────────┘
                    ▼
              SOCIAL NORMS
                    │
                    ▼
             TRAVEL BEHAVIOUR
                    │
                    ▼
             PUBLIC TRANSPORT
                ADOPTION
                    │
                    ▼
             EMISSIONS / HEALTH
                    │
                    ▼
            ENVIRONMENTAL DATA
                    │
                    └──────────► RESEARCH

That is a genuine closed-loop sustainability system.

---

12. Your "quantum sustainability" contribution

I think this is where your synthesis becomes much clearer.

Your book does not need to claim that quantum mechanics explains human behaviour.

Instead, your contribution can be:

Quantum Sustainability = applying quantum and quantum-inspired computational approaches to complex, interconnected sustainability decision problems, while benchmarking them against established classical methods.

The Singapore public-transport case then becomes a demonstration framework.

Your eight-domain idea could eventually extend:

Transport → Energy → Climate → Supply Chain → Materials → Environment → Social Networks → Peace/Conflict

But public transport is an excellent starting case because it naturally combines:

people + infrastructure + economics + behaviour + environment + government + technology + networks.

---

13. Ethics must be a separate layer

I strongly agree with the point you quoted about manipulation risk.

Your system could theoretically discover:

"Which people are most psychologically susceptible to changing transport behaviour?"

That creates a serious ethical issue.

Therefore:

\[
Optimisation
\neq
Manipulation
\]

The objective should be:

Acceptable

Help citizens make informed sustainable choices.

Potentially problematic

Exploit psychological vulnerabilities to force behaviour.

So the framework should include:

Consent → privacy → transparency → explainability → fairness → opt-out → independent ethics review.

For example, don't optimise simply:

\[
\max Adoption
\]

Instead:

\[
\max
\left(
SustainabilityImpact
+
CitizenWelfare
+
SystemEfficiency
\right)
\]

subject to:

\[
Privacy \geq P_{min}
\]

\[
Fairness \geq F_{min}
\]

\[
Autonomy \geq A_{min}
\]

\[
ManipulationRisk \leq R_{max}
\]

This makes your framework substantially more mature.

---

14. A full-scale Singapore research programme

I would structure it into five phases.

Phase 1 — Observe

Collect:

• ridership
• station flows
• bus usage
• travel times
• reliability
• weather
• demographics
• land-use characteristics
• transport accessibility
• environmental indicators.

LTA already publishes transport statistics and datasets that could form part of the empirical foundation.

Phase 2 — Understand

Use behavioural research to identify:

Why does Citizen A choose car while Citizen B chooses MRT?

Phase 3 — Network

Map:

citizen ↔ family ↔ workplace ↔ community ↔ infrastructure ↔ government ↔ technology ↔ environmental organisations.

Phase 4 — Optimise

Test:

classical → quantum-inspired → quantum

on intervention allocation.

Phase 5 — Field experiment

Deploy selected interventions in controlled communities and measure:

\[
Before \rightarrow Intervention \rightarrow After
\]

against appropriate control groups.

---

15. The most important research metric

I would not make public-transport ridership alone the main dependent variable.

Create a broader:

Community Sustainable Mobility Index — CSMI

For example:

\[
CSMI =
w_1 PT
+w_2 Reliability
+w_3 Accessibility
+w_4 EmissionReduction
+w_5 Equity
+w_6 CitizenSatisfaction
+w_7 BehaviourPersistence
\]

The last variable is particularly important.

Someone taking MRT once because of a $2 reward is not the same as someone who changes their long-term habit.

Therefore:

\[
ShortTermAdoption
\neq
SustainedBehaviourChange
\]

That distinction would make your research much stronger.

---

16. Why this fits your original book

Your original proposition can therefore evolve from:

"Quantum computing can help sustainability."

into something much more academically defensible:

"Sustainability challenges are interconnected socio-technical network problems. Quantum and quantum-inspired computation may provide additional tools for selected optimisation problems within these networks, but their value must be established empirically against classical alternatives."

And Singapore becomes the living laboratory.

The country is already moving in this direction: the current refresh of the Land Transport Master Plan explicitly recognises demographic change, climate/weather pressures and advances in AI and autonomous vehicles, while seeking public and stakeholder input.

The Green Plan also explicitly connects transport behaviour with environmental sustainability and sets measurable public-transport targets.

In one sentence

Your strongest research proposition is not "quantum will make Singaporeans take more public transport"; it is "can a multi-layer citizen–behaviour–institution–technology network model discover and optimise sustainable transport interventions, and does quantum/quantum-inspired computation add measurable value over classical approaches?"

That is a research hypothesis that can actually be tested, rather than a claim that needs to be assumed.