Passenger-Freight Shared Mobility

• PFSM is an integrated transport system that dynamically allocates shared vehicle capacity for both passengers and freight, aiming to optimize operational efficiency and reduce costs.
• Advanced mathematical models and metaheuristic algorithms underpin PFSM research, ensuring feasible profit allocations and robust real-time scheduling in complex transport networks.
• Empirical studies show that PFSM can boost revenue, halve fleet sizes, and significantly cut carbon emissions, making it a viable solution for urban and rural mobility challenges.

Passenger-Freight Shared Mobility (PFSM) refers to the coordinated use of shared mobility resources—vehicles, scheduling, and infrastructure—for both passenger and freight transport in integrated systems. By enabling dynamic allocation of vehicle capacity between passengers and goods, PFSM aims to improve operational efficiency, reduce costs, support sustainability targets, and address challenges facing conventional mobility and logistics services, especially in urban and urban-rural transport networks. Recent research has formalized PFSM through robust optimization frameworks, agent-based models, and operational algorithms, demonstrating significant synergies in revenue, resource usage, and environmental impact.

1. Mathematical and Market Design Foundations

PFSM builds on advanced mathematical formulations developed for shared mobility markets. The assignment of passengers and freight to vehicles is commonly framed as a linear programming (LP) or mixed-integer programming (MIP) problem, where the system planner seeks to maximize social welfare subject to capacity constraints and heterogeneous utility functions. For example, the utility of traveler $i$ assigned to vehicle $j$ is given by $u_i(a_{ij}) = v_i(a_{ij}) - t_i(a_{ij})$, with $v_i(a_{ij})$ representing satisfaction and $t_i(a_{ij})$ representing payment (Chremos et al., 2020). Capacity constraints—$\sum_{i\in I} a_{ij} \leq \epsilon_{j}$ for each vehicle $j$—govern both passenger and freight demands.

Stability conditions for profit allocation ensure that neither travelers nor vehicles have incentives to deviate from the provided assignment. In PFSM extensions, both passenger and freight customer types are included, each with distinct utilities and cost structures. The market solution requires that profit allocations are feasible and stable for all agent types, defined mathematically by inequalities that guarantee no coalition benefits from reassignment.

2. Optimization Algorithms and Metaheuristics

PFSM solution schemes span exact algorithms and metaheuristics tailored to high-dimensional, combinatorial settings. Bilevel optimization models govern the collaborative scheduling of multi-type vehicles and dynamic allocation of passenger/freight capacities (Wu et al., 24 Sep 2025). The bilevel structure typically features:

• Upper-level: Vehicle type selection and capacity allocation maximizing economic profit.
• Lower-level: Scheduling and routing minimizing total passenger travel time, incorporating stochastic traffic conditions (e.g. stochastic BPR functions for travel times).

The objective function commonly integrates revenue ($E_1$ for passenger fares and $E_t$ for freight fees) minus running and fixed costs as

$$\max Z = E_1 + E_t - C_{km} - C_{idle} - C_{fix} - C_{toll}$$

To address the premature convergence of standard metaheuristics in solving large-scale, mixed discrete-continuous PFSM problems, the improved jellyfish search algorithm (IJS) has been introduced. The IJS employs tent chaotic maps for population initialization, hybridized movement strategies, and integrates Lévy flight and differential evolution (DE) for enhanced global search capability and solution diversity. Entropy-based weighting methods balance economic, efficiency, and environmental objectives.

3. Integrated Operational Strategies and Service Design

Agent-based simulation models have been adapted for PFSM to capture the interactions among vehicles, passengers, and freight items (Fehn et al., 2022, Bhatnagar et al., 2022). These models:

• Encode service rules via statecharts and discrete events reflecting real-world trip dynamics.
• Implement agent-based scheduling and optimization (e.g., reduction of unserved requests/lost demand).
• Support joint mobility options such as dynamic vehicle cycling between passenger and freight assignments or intermodal transfers.

Heuristic parcel assignment strategies for integrating freight with passenger services include the Combined Decoupled Parcel Assignment (CDPA), Subsequent Decoupled Parcel Assignment (SDPA), and Subsequent Coupled Parcel Assignment (SCPA) (Fehn et al., 2022). These algorithms balance additional distance, delay, and customer inconvenience, typically governed by detour thresholds $\tau_{th}$.

Rolling horizon algorithms and metaheuristics (e.g., Adaptive Large Neighborhood Search) dynamically manage trip planning and vehicle routing under changing passenger and freight demands (Zhang et al., 20 Feb 2025). Such frameworks support near-real-time optimization, ensuring demand fluctuations, schedule flexibility, and heterogeneous preferences are accommodated.

4. Economic, Efficiency, and Environmental Impacts

Empirical studies citing PFSM implementations demonstrate the following outcomes:

Metric	PFSM Improvement	Source
Operating revenue	Increase by up to 328.45%	(Wu et al., 24 Sep 2025)
Carbon emissions (freight)	Decrease by 19.12 tons/year	(Wu et al., 24 Sep 2025)
Passenger travel time	Moderate increase (e.g. 19.46%), deemed acceptable	(Wu et al., 24 Sep 2025)
Fleet kilometers (integrated operations)	Up to 48% reduction (parcel delivery segment)	(Fehn et al., 2022)
Fleet size requirements	Reduction by an average of 35% (multi-purpose)	(Hatzenbühler et al., 2022)
Vehicle trip duration (multi-purpose fleet)	Reduction by 33%	(Hatzenbühler et al., 2022)
Total distance travelled	Reduction by 16%	(Hatzenbühler et al., 2022)

The reallocation of vehicle resources and the dynamic assignment of passenger and freight loads supports greater operational efficiency, higher vehicle utilization, and lower overall costs. Modular vehicle configurations—enabled by module switching at depots—allow fleets to flexibly serve both market segments, which is critical in infrastructure-constrained urban and rural settings.

Sensitivity analyses in PFSM research identify optimal capacity allocation thresholds (often 40–60% passenger capacity recommended) and service time constraints that maximize profit without sacrificing user experience.

Environmental analyses highlight the potential of PFSM to drastically reduce both vehicle kilometers and carbon emissions by replacing dedicated freight and passenger runs with joint services. Decoupled fleet operations are consistently outperformed in multi-objective optimization studies.

5. Behavioral, Market, and Preference Integration

Contemporary PFSM frameworks integrate passenger and shipper behavioral decision-making via utility functions informed by real-world surveys or advanced choice models (e.g., Cumulative Prospect Theory for passenger risk attitudes (Nair et al., 2021)). These approaches account for heterogeneity in risk aversion, time and cost sensitivities, and acceptance probabilities for dynamic tariffs.

Preference-based optimization frameworks allow for the direct incorporation of stated preferences, yielding trip assignments that holistically maximize social welfare across diverse user groups (Zhang et al., 20 Feb 2025). Dynamic pricing schemes, subject to nonlinear sensitivity analysis, modulate tariffs to simultaneously maintain service attractiveness and profitability in mixed passenger-freight contexts.

Stability considerations ensure no agent (passenger, driver, shipper) has an incentive to deviate from market assignments, preserving equilibrium and long-term operational viability (Chremos et al., 2020).

6. Policy and Structural Implications

PFSM market design and optimization analyses have direct consequences for transport planning and policy:

• Integrated scheduling and resource allocation frameworks can guide investments in modular vehicle platforms and supporting depot infrastructure.
• Integer-based equilibrium models support evaluation of subsidy/tolling schemes, planning for intermodal transfer capacity, and regulatory compliance for mixed-use vehicle fleets (Kadem et al., 1 Feb 2024).
• Studies quantify the price of anarchy arising from unmanaged user equilibrium, motivating policy interventions to align individual incentives with societal objectives (e.g., system-optimal routing and dynamic incentive structures).
• Agent-based and simulation-optimization models aid practitioners and policymakers in quantifying the trade-offs and potential gains from PFSM adoption in various regional and network contexts.

7. Future Directions and Research Challenges

Current PFSM research suggests several avenues for continued paper:

• Further refinement of heuristic algorithms for operational planning, including asymmetric detour thresholds and temporal grouping of similar freight deliveries.
• Extension of agent-based simulators to model richer agent states (e.g., reverse logistics, urgent freight, and variable passenger time windows).
• Operational strategies for automated and connected vehicles in PFSM, leveraging improved utilization rates and cost models for autonomous trucks and shared shuttles (Jungblut et al., 17 Aug 2024).
• Urban-rural system integration, focusing on last-mile logistics and rural accessibility, with bilevel optimization and environmental impact analysis (Wu et al., 24 Sep 2025).
• Deeper exploration of joint mobility partnerships between public transit agencies, logistics companies, and rideshare platforms, including revenue-sharing, bundled fares, and dynamic rebalancing algorithms.

Technical implementation remains contingent on robust real-time data integration and careful calibration to capture the stochasticity and non-stationarity of both passenger and freight markets. Regulatory and safety considerations, especially in the context of passenger-freight mixing and urban network constraints, require further specification and harmonization.

Conclusion

Passenger-Freight Shared Mobility (PFSM) represents an interdisciplinary fusion of transportation, operations research, behavioral economics, and systems engineering. It leverages advanced optimization methodologies, behavioral models, and intelligent service design to align the interests of diverse mobility stakeholders, driving efficiency, sustainability, and economic performance beyond the limitations of traditional, segregated transport operations. Recent studies have validated the model’s effectiveness through mathematical programming, simulation-based experiments, and policy evaluation, but continued research into algorithmic refinement, agent modeling, and policy integration is essential to fully realize its potential in complex, evolving mobility ecosystems.