Developed solutions and services

MODALSHIFT aims to provide an e-subscription device for vulnerable citizens using public transport. The device is going to facilitate their access to public transport to monitor their trips and provide added value information to guide them in their intermodal journeys. The device is going to help users by enabling them to buy tickets online and by communicating with buses and the surrounding vehicles improving the security of vulnerable users with alerts to nearby connected car users.

This e-subscription and tracking device will be developed by NGS using Bluetooth Low Energy technology and based on the tracking platform offered by GGI. It is going to manage user-based data collection through the OBU. Therefore, NGS will customize their IoT scalable sensor to integrate the relevant payment and connectivity functionalities. From the operator perspective, by integrating the IoT data into the Data Space (S6) and including it the MTM tool (S10) this service can also be explored as a source of live demand data with associated benefits.

For the development of the shared municipality-owned cargo-bike service in the VARNA case study, MODALSHIFT will connect the T&T&M technology for data collection into an app, giving visibility to local companies on how many shared bikes are currently in use or allowing them to book a time slot, using NGS’s 5G battery-less device. For privacy reasons, the bikes’ position is not available when in use by other stakeholders. The data transmitted by the IoT devices provided by NGS and from the environment will be converted into the GS1 EPCIS 2.0 standard by NGS and made available to GGI, which will process and display them within its platform. GGI will improve its T&T&M portal with functionalities to meet the requirements of the VARNA CS for city-owned cargo-bike fleet management, with self-registration of local businesses, and overview of the municipality about the users. Registered users can see cargo-bike availability and location, reserve them, and track their usage by operators. No location is shown when the vehicles are in use by other business or put offline by the city for servicing. The municipality can extract anonymized information about the locations, and have a report on the utilisation of vehicles, payment and potential misuses (e.g. late return, damages). This data shall also be available via the Data Space (S6) and support analytics and patterns via de MTM tool (S10). 

This e-subscription and tracking device will be developed by NGS using Bluetooth Low Energy technology and based on the tracking platform offered by GGI. It is going to manage user-based data collection through the OBU. Therefore, NGS will customize their IoT scalable sensor to integrate the relevant payment and connectivity functionalities. From the operator perspective, by integrating the IoT data into the Data Space (S6) and including it the MTM tool (S10) this service can also be explored as a source of live demand data with associated benefits.

Among the number of devices connected to the multimodal collaborative traffic and network management platform, there will be a dynamic fleet management and drivers’ app based on real-time optimisation algorithms. The mobile app for logistics fleet management developed by GGI will be further improved since for now it is limited to the management of the fleet of a single user for cargo bike urban deliveries. MODALSHIFT aims at designing a tool suitable for efficient and sustainable urban parcel distribution, based on the GGI app in deliveries. The app allows the tracking of the shared vehicles by the registered companies when in use, while focusing on the privacy requirements to avoid a company tracking competitors. These privacy requirements are equally implemented in the Data Space (S6) and MTM Tool (S10). 

Imec has developed expertise on route travel time prediction models capable of improving the accuracy of traffic state forecasting through additional components like road / time features or weather information applicable on both public transport and logistics traffic as well as prediction on road network, particularly on segments where induction loops, cameras, and/or map-matched vehicle trajectories are available. These predictive models are also capable of forecasting the freight demand (e.g., volume of incoming cargo) and capacity (e.g., workforce capacity by men hours) at logistics hubs useful for optimization of transport routing. MODALSHIFT will further develop the current network state forecasting and extend it to rail network application for the Italian case study. Through specific transfer learning techniques and preprocessing strategies, a sound use of AI in a multimodal framework and the identification of relevant machine learning based techniques lack of data will be reduced and exploitation of multi-source data will be improved. The Road & Rail traffic forecasting tool is meant to improve the accuracy of traffic forecast and improve the capacities of anomaly detection algorithms to compare forecasts and on-ground situation. This tool will be accessible via the Multimodal Traffic Management Tool (S10) where live data and forecasts can be explored, used for alerts and combined with other data sources and visualizations.

The concept of a smart box will be adapted to Capacity-as-a-Service applications through the building of a prototype smart locker, containing several smart boxes that can be swiftly loaded/unloaded between the transport mean and the stop. The smart box acts as a service for logistic unit consolidation and characterisation, enabling the buying of a shipment and collecting data, with Track&Trace&Monitoring (T&T&M). The idea is that smart boxes, once filled by couriers, are loaded into the smart locker installed inside the public transport vehicle. When the vehicle reaches the designated stop where a second smart locker is installed, the system will automatically transfer the smart box—using, for example, a robotic arm—from the locker inside the vehicle to the locker at the stop.

The smart box exchange will be scheduled and managed at the dynamic logistics fleet management scale with synchromodality strategies to maintain the public transport network performance. Parcel retrieval is done with an NFC connection. NGS develops prototypes of the smart boxes and the smart lockers following ISO sizing standards. The Logistics-as-a-Service approach will embed services like logistics units’ characterisation (weight, size, number of items), T&T&M, and the possibility of buying a shipment. Designing this smart box enabling Capacity-as-a-service is a solution for reducing traffic, through the use of public transport for logistics, while adding a new data source helping V2X operations and traffic forecasting. This new data source will be made available in the Data Space (S6) and, through the MTM Tool (S10), will be explored in terms of synchromodal optimisation potential.

In order to tackle current data standard and format harmonisation problems and foster data space security in the logistics and mobility sector through a multimodal perspective, MODALSHIFT will design and implement a new data space architecture for the 3 Case Studies integrating mobility, logistics and transport infrastructures. Based on the learnings of ITA gathered in HE GEMINI (2023 – 2026) and best practices from EMDS, Digital Transport and Logistics Forum and EU data strategy (e.g. supported data harmonisation, addressing the requirements of various data types, minimising redundancies by streamlining knowledge exchange, etc.), the project ambitions to convert data in open REST APIs into the Data Spaces, in line with the EU Directive on open data and the re-use of public sector information. In this process, FIWARE Data Spaces is the proposed Data Space Connector and framework intended for implementation. The aim is to make the data spaces completely compatible with GDPR with the anonymised access to datasets from private actors, while ensuring the alignment with stakeholders’ own formats and maturity levels. Moreover, MODALSHIFT will develop standard-based connectors for shared access in the data space for each category of transport stakeholder, facilitating the adoption and customisation to the specificities of stakeholders’ data, and integrating trust-by-design with the anonymisation processes. Moreover, these anonymization processes can also be used to desensitize data that do not contain personal but industrial data. The aim of desensitization is to modify information that could potentially reveal strategic or commercial insights about a company.

By complementing the possibilities offered by anonymization, desensitization can thus also contribute to encouraging local stakeholders to share their data. NGS will support this architecture by providing a GS1 EPCIS-compliant repository and a blockchain-enabled cloud gateway to ensure secure and standardised logistics data exchange. MODALSHIFT will propose practical guidelines for the 3 national/local contexts of Case Studies to support evolutions of data regulations, building on the key directions identified by the MTM cluster (e.g. mandatory open data access) and test phase in these CS. NGS will support this architecture by providing a GS1 EPCIS-compliant repository and a blockchain-enabled cloud gateway to ensure secure and standardised logistics data exchange.

To integrate the stakeholders’ and end users’ perspective in traffic forecasting and optimisation, as well as introduce detailed data on e.g. users’ choices and freight flows, MODALSHIFT is going to introduce agent-based modelling (ABM) into multi-objective decision-making algorithms. Hence, AIT will use separate models to produce homogenous user groups using mobility information types (Dangschat & Millonig, 2023) and build on the available data shared through data spaces.

TELLAE has developed an open-source and flexible agent-based simulator dedicated to transportation issues called Starling in the BPI DIVD Rennes Metropole COMOB project (2020-2021), initially used for shared mobility (carsharing, on-demand PT and bike sharing services). Through MODALSHIFT, TELLAE will improve the Starling open-source core framework, and community extensions, and will further develop this agent-based model to cover more realistic UCs integrating traffic flow congestion, optimisation of regulatory operations, timetable updates, demand models. Additional mobility services like on-demand mobility and several variants that matches needs of vulnerable populations are going to make the UCs even closer to the reality of the transportation ecosystem and improve the ABM accuracy with positive impacts in strategic and management decisions. MODALSHIFT also aims to measure and improve how the design of services actually answers the citizen needs. Relevant results can be integrated in S10 to be available in combination with the full set of available data. 

A variety of incentives have already been studied for shifting the behaviours of transport system users towards low-carbon options such as point reward systems or dynamic pricing for road use, but they lack a multimodal approach. MODALSHIFT ambitions to design dynamic incentives, helping the reduction of traffic congestion by acting on the offer of people and goods transport services and traffic management framework.

These incentives will span three main areas of exploration: pricing of the transport services, information provision, tax, and regulation. The developed incentives will benefit in priority vulnerable groups thanks to the identification of segmented incentives and will be tested by ATOBE in the CS’s multimodal hubs. MODALSHIFT will ensure the viability of the incentives with the study of business models of the transport service provider with more than 2 models studied and developed per CS. A value canvas will be developed for each category of user and stakeholder for designing tools and services tailored to each type of stakeholder needs. To integrate logistics in dynamic incentives for traffic reduction, MODALSHIFT finally aims to identify the opportunities for evolving local regulations in a dynamic perspective with 1 policy white paper and 1 workshop per CS, towards further engaging logistics operators in adapting their delivery planning with adjusted taxes and authorisations. 

TELLAE has developed KITE as a platform bringing complete tools for transport planners to enhance decision-making: from data analysis to what-if simulation, giving access to easy-to-use ABM simulation, simplified and more sophisticated simulation results. MODALSHIFT will enhance the functionalities of the KITE tool through combining data visualisation of raw observed data, enhanced data like synthetic data of population and trips, agent-based simulation results, and adding simplified simulation tools in order to bring together all data around public transport and shared mobility, allowing easy analysis and data visualization. New data connectors to open data platform will be developed to support the adaptation to the three CS countries targeted. As the volumetry of data will increase (OpenStreetMap, etc.), data management processes will be adapted depending on the fit of various architectures and strengthening algorithms will be added as open data often suffer from a lack of quality or too much inconsistencies, notably OpenStreetMap (lack of consistency between open contributions) and public transport data (often errors).

New algorithms for cross analysis between data will be developed with python functions and API services to make available KPIs for territory analysis. In the KITE front end, MODALSHIFT will focus on the simulation integration. Besides Starling, other simulators will be connected to KITE with new widgets in order to change the public transport routes, evaluate demand, etc. 

One key aspect of the MODALSHIFT project is to overcome existing limitations and segmentation in management (and data) in multimodal infrastructures. To directly address this aspect, ATOBE will build on one of the main results of the TANGENT Horizon 2020 Funded project, the TANGENT Dashboard. This Dashboard relies on a configurable data visualisation engine, coupled with planning and incident management with a great focus on multi-entity cooperation built in. 

The MTM tool will integrate results from WP3, providing combined visualisation and planning, namely of the digital twins resulting from T3.1 “Development and calibration of Digital Twins modelling freight, mobility and network”. For T3.2 Network and traffic forecasting, by integrating S4 as part of the core tool, making forecast data a first-class citizen of the tool. For T3.3 “Agent-based modelling for identifying ‘socially optimal’ multimodal framework” mostly by integrating data from S7. For T3.4 “Dynamic network optimisation coupling predictive and prescriptive analytics” by integrating data from S2 and the broader results of T3.4 (not covered in S2). Finally, for T3.5 “Synchromodality for logistics-mobility integrated services in intermodal nodes” either by integrating the data through the digital twins in T3.1 and/or by integrating T3.5 results into the MTM Tool Planner. 

By building on top of the WP2 Data Space and natively allowing the configuration of dashboards using live and forecast data and providing cooperative planning and management tools supported by the results of WP3 and other WP4 modules, the MTM Tool will be able to tackle several limitations such as a lack of information about the transport network, the limited cooperation between stakeholders, lack of a single system with all relevant real-time data, or inefficient planning and modelling of the infrastructure. 

The MTM Tool is meant as a horizontal platform covering the visualisation and cooperative management needs of the multiple stakeholders involved in the CS. In MODALSHIFT this platform is also combined with a set of tools that cover vertical aspects (namely S1, S2 and S5), providing more domain specific tools and covering innovative operational aspects of the CS.