Smart Infrastructures

Maintaining and renewing our infrastructure will require considerable investment in coming decades. Major changes are needed in our transport and energy systems, particularly to meet environmental challenges. At the same time, advances in information technology provide an opportunity for GERAD members to work at improving the capacity, efficiency and reliability of our infrastructure rather than simply repairing it.

More Data Available

Smart infrastructure aims to improve – often in real time – a service provided to a population of users, using different kinds of available data about their own condition as well as about users themselves. For example, management of a public transport network or a taxi fleet can rely on the precise positioning of vehicles, rapid reporting of incidents, as well as forecasts of traffic and changes in demand made more reliable by location data transmitted by the cell phones of users. More generally, with the development of the Internet of Things, a proliferation of sensors and data sources are becoming available in many areas. GERAD brings together several researchers working to transform and merge this raw data into statistical models than can then be used for decision-making purposes. Such data can also be used for longer-term planning of changes to infrastructure.

Social and Ethical Dimensions

Personal data are the pillar of the idea of smart infrastructure, but there are many legitimate concerns about the practical uses of personal data, since its dissemination may affect privacy. Many cities have developed "open data" programs which do not necessarily apply the state-of-the-art methods that are needed to protect private data. For example, it is well documented that datasets recording the movements of individuals over time are particularly hard to anonymize, although such datasets can frequently be found in open access. Professor Jérôme Le Ny's research group is interested in development of estimation and decision-making models that can use aggregated personal data, while providing formal mathematical guarantees with respect to the protection of data confidentiality (so-called "differential confidentiality" guarantees). This includes developing systems that collect data only for well-defined purposes, and then applying protection methods (aggregation, scrambling, etc.) that are specifically tailored for these purposes in order to limit the impact on statistical accuracy. There are many such applications, and this research work will contribute to strengthening the trust of users in smart infrastructures as well as their consent to providing data that are needed.

Examples of publications:

• Le Ny, J., Differential Privacy for Dynamic Data. SpringerBriefs in Control, Automation and Robotics, Springer, 2020.

• Le Ny, J., Privacy in Network Systems. In Encyclopedia of Systems and Control, J. Baillieul, T. Samad, Editors, Springer, 2021.

• Pelletier, M., Saunier, N., Le Ny, J., Differentially Private Analysis of Transportation Data. In Privacy in Dynamical Systems, F. Farokhi, Editor, pp. 131-155, Springer, 2020.

Planning Operations in Public Transport

All large cities have a public transport network that comprises, among others, a bus service. This infrastructure offers to the population the possibility to travel between the various neighbourhoods of the city in an economical and environmentally friendly fashion. Public transport companies are to a large extent subsidized by governments and must therefore offer good-quality service while avoiding excessive operating costs. Consequently, they rely on decision-making software to optimize operational planning.

Given that task complexity (for example, the Société de transport de Montréal offers more than 17,000 bus trips per day, distributed among 225 bus routes), planning for a bus network is typically done by steps. Network planning includes development of: i) bus routes; ii) trip schedules for each route; iii) bus schedules indicating the sequence of trips to perform for each bus; iv) daily work schedules specifying the sequences of trips an anonymous driver must cover; and v) drivers' schedules assigning work duties and days off to each driver for the next month. The problems in steps i) and ii) are at the strategic/tactical level and relate to maximizing the quality of service while respecting some global constraints on resource availability. On the other hand, the problems in steps iii) to v) are operational ones that relate to offering at least cost the service established in the previous steps while taking into account numerous practical constraints such as those stemming from the drivers' collective agreement.

For several decades now, GERAD's researchers have realized research projects on the operational problems of steps iii) to v). Most of these works have been conducted in collaboration with GIRO company, which is the world leader in the commercialization of optimization software for public transport. To solve the driver duty scheduling problem, François Soumis and Jacques Desrosiers have developed at the end of the 1980s, a column generation algorithm, called Gencol, that allows to select the best set of work duties to perform among a huge number of possible duties without having to enumerate them all. This algorithm, which is still in use at GIRO, has been enhanced over the years with new advancements made at the GERAD, namely, the dynamic constraint aggregation technique designed by Issmail El Hallaoui, François Soumis and Guy Desaulniers. More recently, GIRO has decided to also apply column generation for solving the bus scheduling problem in order to handle the recharging constraints of the electric buses. In this regard, a new collaboration with Guy Desaulniers has permitted to study different variants of this problem, including that considering the possibility to slighlty shift the trip start times. Finally, Guy Desaulniers, Andrea Lodi and François Soumis are teaming up to integrate statistical and machine learning methods in the optimization algorithms for public transit.

Scaling Up Battery Swapping Services in Cities

Electrification of urban transportation is at the heart of burgeoning smart-city transformations. Electric vehicles run on batteries and electric motors, and therefore offer the prospect of benefits such as reducing carbon emissions. These anticipated benefits have pushed governments around the world to aggressively incentivize the adoption of electric vehicles. For example, the UK and France plan to ban the sale of new internal-combustion vehicles by 2035 and 2040, respectively. Chinese megacities ration more electric vehicle sales than fossil-fuel car sales by capping the numbers of issued licence plates. Given these trends, over 500 million passenger vehicles on the road, or 30% of the entire fleet, are expected to be electric by 2040.

As electric vehicles thrive, battery swapping is reviving. Battery swapping is an alternative to plugging in cars for battery charging: it refers to refuelling an electric vehicle by replacing the depleted battery on board with a charged one. The revival of battery swapping is driven by its three advantages over plug-in charging: 1) Speed: The swap process takes only 3-5 minutes, whereas using even Telsa's supercharger takes more than 30 minutes to charge a battery to 80%. 2) Compactness: The short service time allows a swapping station to occupy much less space than a plug-in charging station in order to achieve the same service level. 3) Safety: Compared with vehicle owners, service providers can more efficiently charge and maintain batteries. Therefore, battery swapping is widely believed likely to prevail, especially in large cities.

Nevertheless, electric vehicle companies and municipalities are now facing major challenges when they try to scale up battery swapping services in cities:

• Demand Uncertainty and Service Proximity: The infrastructure deployment needs to be dense so that random swapping demands are able to access a swapping station without much detour. The resulting urban swapping station networks are highly decentralized and thus difficult to operate.
• Battery Availability: Swapping stations also need to ensure a high availability of charged batteries in stock to satisfy swapping demands and to avoid queueing. This operational challenge is exacerbated by the decentralized layout of swapping stations, since swapping demands are more variable when disaggregated than when pooled. Consequently, the service provider has to build up massive battery inventories, which are expensive and environmentally detrimental.
• Grid Accessibility: Finally, charging depleted batteries at swapping stations may overload or even destabilize local low-voltage power distribution grids. These "last-mile" grids were often built without allocating enough capacity for electric vehicle charging. Upgrading existing distribution grids would be prohibitive, if not infeasible.

For years, GERAD's researchers have been working on operations research problems concerning greening our urban transportation systems under uncertainty. In particular, Wei Qi and his collaborators within and outside of GERAD have been working on providing deeper understanding of how to scale up citywide battery swapping services in order to cope with the aforementioned challenges. For example, these researchers are examining a "swap locally, charge centrally" network setting where batteries are locally swapped at decentralized swapping stations, and transported to and charged at more centralized charging stations that are connected to grids of sufficient capacity (which are typically of higher-voltage levels or near a substation). They have built models that analytically characterize the intertwined and stochastic operations of stocking, swapping, charging, and circulating batteries between a centralized charging station and decentralized swapping stations. They have also developed a new algorithmic framework combining constraint-generation and parameter-search techniques to solve intricate joint infrastructure planning and operations problems.

Reference:

• Qi, W., Zhang, Y., Zhang, N., Scaling Up Battery Swapping Services in Cities (June 19, 2020).

The Case of Telecommunications Infrastructure

As the world becomes "smart", with "smart cities", "smart grids", "smart buildings", etc., it is also becoming increasingly reliant on its telecommunications infrastructure. To answer the needs of smart applications, the underlying infrastructure must be extremely reliable, sustainable and highly adaptable to the system's changing conditions. Current and forthcoming smart systems increasingly require telecommunication networks that are ubiquitous, all pervasive, highly performant and transparent to users. This smoothness in network operation requires a complex, large scale and highly heterogeneous telecommunications infrastructure. Wireless, optical, satellite, aerial, computing and storage technology must efficiently work together to provide the availability and the response time that is required by smart applications. Networks need to rearrange their resource offering in the nick of the time, and that is only possible with the increased use of virtualization to define network functions, alongside a form of centralized network intelligence based on the concept of Software Defined Networking (SDN). Thus, telecommunications infrastructure itself must be the smartest, for all the other smart systems relying on it to function properly in real time.

An example of such infrastructure is currently provided by the deployment of 5G in large cities that may set virtual networks, called "slices", to different smart systems, such as intelligent transportation, connected vehicles or telemedicine, each having applications with different real-time performance requirements. The challenge for operators is to ensure that those systems are always available, and that the response time provided by the infrastructure is suitable for each smart application. To help in this task, a citywide 4G/5G simulator has been developed at GERAD by Professor Brunilde Sansò's research group. The large-scale discrete event simulator recreates equipment specifications normalized by the telecommunications standards entities. It also uses real telecommunications and city infrastructure data to assess the response time of key applications and to detect problems in the network. One of the major challenges tackled by the research group is the large scale modelling of the system that may contain hundreds of base stations operating at the millisecond level in a citywide networking mode. Another challenge is the mapping of the citywide application into the simulated infrastructure. From the time-scale standpoint, the mapping implies assessing the application performance, in minutes or hours while simulating milliseconds, which involves advanced statistical and machine learning modelling. Another challenge arises from mobility applications, such as V2V or ITS that, in their path towards their destination, make use of different telecommunications equipment. Thus, the mapping needs advanced algorithms for distributed computations. Finally, machine learning methods are put in place to assess the performance of the applications and to identify network problems, such as dormant cells or denial of service attacks. An on-line version of the simulator can be accessed here.

A final thought on real-time decision-making and telecommunication infrastructure is that the "smartness", reliability and high availability of such infrastructure comes with a hidden price: increased energy consumption and environmental impact of data centres and network components. For years, Professor Sansò's group has been interested in ways of guaranteeing network reliability and availability while reducing energy consumption and environmental impact. Among others, the intelligent real-time operation of data centres, the design and operation of wireless access with energy constraints and the study of networks fed by solar energy. The next step currently being developed as an extension of the above-mentioned citywide simulator is how to integrate unreliable energy sources to insure real-time response of smart applications even in challenged and catastrophic conditions.

References:

• Manzanilla-Salazar, O.G., Malandra, F., Mellah, H., Wette, C., Sansò, B., A Machine Learning Framework for Sleeping Cell Detection in a Smart-City IoT Telecommunications Infrastructure. IEEE Access, 8, 61213-61225, 2020.

• Seyedi, Y., Karimi, H., Wette, C., Sansò, B., A New Approach to Reliability Assessment and Improvement of Synchrophasor Communications in Smart Grids. IEEE Transactions on Smart Grid, 11(5), 4415-4426, 2020.

• D'Amours, M., Girard, A., Sansò, B., Planning Solar in Energy-managed Cellular Networks. IEEE Access, 6, 65212-65226, 2018.

• Malandra, F., Chiquette, L.O., Lafontaine-Bedard, L.P., Sansò, B., Traffic characterization and LTE performance analysis for M2M communications in smart cities. Pervasive and Mobile Computing, 48, 59-68, 2018.

• Larumbe, F., Sansò, B., A Tabu Search Algorithm for the Location of Data Centers and Software Components in Green Cloud Computing Networks. IEEE Transactions of Cloud Computing, 1(1), 22-35, 2013.

• Boiardi, S., Capone, A., Sansò, B., Radio planning of energy-aware cellular networks. Computer Networks, 57(13), 2564-2577, 2013.