An IoT approach for optimizing routing and safety?

Abstract

This paper presents an IoT approach that aims to provide routing optimization and navigation as well as to improve safety of commercial vehicles. The proposed approach utilizes static information and data to generate an e cient overall delivery sequence of customers for the whole eet of the company and real-time tra c information to assist drivers in making more e ciently a particular point-to-point trip. With regard to safety, improvements could simply be achieved by easing driver mental workload and reducing their exposure to higher-risk roads of heavy tra c.

Keywords

• urban goods movement
• routing optimization and naviga- tion
• safety improvement

-

As markets tend to become increasingly competitive, e cient urban goods movement turns out essential and necessary to expand pro t margins for companies and improve customer services. As such, companies plan, implement and control their ows of goods utilizing related information from the point of origin to the point of consumption in order to satisfy customer requirements [ 1 ] and a possible area of improvement is determined in the context of vehicle management and monitoring. The development of modern technologies such as Internet of Things (IoT), Vehicle to Infrastructure (V2I) and Vehicle to Vehicle (V2V) [ 2 ] communication constitute automatic routing and navigation a feasible approach ? This research has been co- nanced by the European Regional Development Fund of the European Union and Greek national funds through the Operational Program Competitiveness, Entrepreneurship and Innovation, under the call RESEARCH { CREATE { INNOVATE (project code: T1EDK - 04012). beyond the common practices for e cient and safer distribution of urban goods [ 3 ].

The high frequency of deliveries in the urban environment often lead to substantial negative externalities such as increased congestion, energy consumption and other safety impacts. In other words, routing and navigation constitute complicated processes for companies that try not only to serve their customers minimizing their travel costs but also to minimize fuel consumption and equipment repairs. Technological advances could o er improved services aiming to safety, capable to monitor real-time road conditions and communicate them with drivers in order to avoid roads of high congestion and risk. The e ective planning and management of delivery schemes may impose a signi cant impact on current distribution operations optimizing the logistics of the produce supply chain thus minimizing several of the relevant externalities.

The aforementioned considerations form the background of this paper, which aims to develop an IoT approach that utilizes V2I communication to optimize vehicle navigation with emphasis on safety. The availability of real-time information provided by modern technology (i.e. tra c cameras, tra c sensors etc.) is considered to provide dynamic navigation that avoid exposure to higher-risk roads that are congested with tra c. The main focus is given on a modi cation of the well-known Vehicle Routing Problem with Time Windows (VRPTW) [ 4 ] in which vehicles communicate in a wireless manner with tra c lights to allow e cient road navigation from a point A to a point B based on the dynamic tra c volume of the road network (i.e. queues, waiting time).

The remainder of the paper is structured as follows: section 2 describes the proposed IoT approach for e cient and safe routing optimization and navigation of commercial vehicles utilizing modern technologies and then, section 3 presents a discussion of the results obtained of a real case study. Finally, in section 4 conclusions are drawn. 2

IoT approach

This paper presents an IoT approach for e cient routing optimization and navigation of commercial vehicles based on V2I communication in an attempt to avoid tra c and enhance safety. The proposed solution (Fig. 1) consists of two main phases: (1) routing phase and (2) navigation phase.

Routing phase aims to provide an e cient overall delivery sequence of customers for the whole eet of the company taking into consideration customer's information about their location, earliest and latest time windows, their service times and demand. The VRPTW is used to model a variety of aspects for distributing goods to customers based on static information and a metaheuristic algorithm is utilized based on a memory-based Tabu Search [ 5 ] which is enhanced with a probabilistic mechanism similar to greedy randomized adaptive search procedures [ 6 ] and [ 7 ] to further improve the solutions utilizing a mechanism for properly switching between local search and modifying the current solution, ensuring that the algorithm balance between intensi cation and di erentiation. 2 Copyright c 2020 for this paper by its authors.Use permitted under Creative Commons License Attribution 4.0 International (CC BY 4.0).

On the other hand, navigation phase aims to support drivers to follow roads without tra c and avoid high risk roads. This diversion practice enhances safety as guides drivers to safer road conditions and reduces crash risks. For this reason, modern road networks utilize modern technology (i.e. tra c cameras, tra c sensors etc.) to calculate travel times and the number of vehicles on a road [ 8 ]. Based on these collected real-time information and data, a backpressure algorithm [ 9 ] is implemented to calculate for each outgoing road a backpressure weight which represents the localized queue and link state information [ 10 ] and [ 11 ]. It acts dynamically based on the network information provided by tra c cameras following a dynamic update process. In particular, for each vehicle a weight is calculated based on the localized queue (i.e. queue backpressure) and link state information (i.e. backlogs at junctions). 3

Empirical study

An empirical study is executed following the proposed IoT approach and a real data set is provided by a transportation company based in Thessaloniki city in Copyright c 2020 for this paper by its authors.Use permitted under Creative 3 Commons License Attribution 4.0 International (CC BY 4.0).

Greece where there are blue-tooth sensors and optical light cameras in infraredsensors. Given a set of depot-returning capacitated vehicles, each customer is visited only once by exactly one vehicle in order to receive products, as shipments are initially located at the depot. The data set contains 35 customers and their addresses (by latitude and longitude), the corresponding time windows and demands.

The spatial distribution of customers is depicted in Fig. 2 and Euclidean distances are calculated considering a constant speed of the vehicle equal to 50 km/h. The service time varies for each customer and depends on customer demand and the time for unloading. The depot is open from 06:00 to 16:00, and the maximum route duration should be up to 8 h.

The aim of this study is to observe and measure the impact of the proposed IoT approach in practice. The routing phase generates the delivery sequence of customers for the whole eet of the company minimizing the total vehicles used and the total traveled distance. Table 1 summarizes the results obtained on the empirical study and the CNR and CTD stand for cumulative number of routes and cumulative traveled distance (without considering tra c). 4 Copyright c 2020 for this paper by its authors.Use permitted under Creative Commons License Attribution 4.0 International (CC BY 4.0).

The navigation phase utilizes dynamic information and aims to support drivers for following the path with lower risk and avoid tra c. Fig. 3 depicts the nal solution obtained for this case study and table 2 summarizes the results derived in practice. The results of the navigation phase appears a weighting factor of 1.43 regarding the distance metric [ 12 ] comparing to the routing phase which means a 30% increase. This indicates the e ciency of the proposed IoT approach and allows determining the qualitative feature of planning for real conditions. Trends in advanced technologies render the development of more sophisticated solutions that allow e cient routing optimization and navigation of commercial vehicles with emphasis on safety. This paper presents how an IoT approach could guide vehicles based on real-time information achieving cost e ciency in terms of travel time and improving safety. The proposed approach is based on the communication among vehicles and infrastructure which is a key for improving transportation process of companies delivering goods in urban areas. This allows dynamic updates on navigation phase comparing to the state of practice Google Maps that requires more time to gain information for an incident or increased tra c presenting a more periodic update process. To this end, this paper presents the applicability and the importance of the proposed IoT approach for advanced management of city logistics and especially in servicing customers in case of emergency due to unexpected events.

Copyright c 2020 for this paper by its authors.Use permitted under Creative 5 Commons License Attribution 4.0 International (CC BY 4.0).

References

1. Simchi-Levi , D. , Wu , S. D. , & Shen , Z. J. M.: Handbook of quantitative supply chain analysis: modeling in the e-business era (Vol. 74 ). Springer Science & Business Media ( 2004 )

2. Miller , J. : Vehicle-to-vehicle-to-infrastructure (V2V2I) intelligent transportation system architecture . In 2008 IEEE intelligent vehicles symposium, IEEE , vol. 1 , (pp. 715 - 720 ). IEEE.

3. Peng , T. , Su , L. , Zhang , R., Guan , Z. , Zhao , H. , Qiu , Z. , & Xu , H. A new safe lane-change trajectory model and collision avoidance control method for automatic driving vehicles . Expert Systems with Applications , vol. 141 , 112953 ( 2020 )

4. Gendreau , M. , and C. D. Tarantilis . Solving Large-Scale Vehicle Routing Problems with Time Windows: The State-of-the- Art . Technical Report CIRRELT-2010-04 . Universite de Montreal, Montreal ( 2010 )

5. Glover , F. : Tabu search|part I . ORSA Journal on computing 1(3) , pp. 190 { 206 ( 1989 )

6. Feo , T. and Resende , M.G.C. : Greedy randomized adaptive search procedures . Journal of Global Optimization 6 , pp. 109 - { 154 ( 1995 )

7. Tarantilis , C.D. , A. Anagnostopoulou and P.P. Repoussis . Adaptive Path Relinking for Vehicle Routing and Scheduling Problems with Product Returns . Transportation Science 47 ( 3 ), pp. 356 { 379 ( 2013 ).

6 Copyright c 2020 for this paper by its authors . Use permitted under Creative Commons License Attribution 4.0 International (CC BY 4 .0).

8. Kopetz , H.: Internet of Things . In: (eds.) Real-time Systems , edited by H. Kopetz, pp. 307 { 323 . Boston, MA: Springer ( 2011 ).

9. Spyrou , E. D. and Mitrakos , D. : A Backpressure Framework Applied to Road Trafc Routing for Electric Vehicles . In: (eds.) Proceedings of the Sixth International Symposium on Business Modeling and Software Design 2016 , BMSD , vol. 1 , pp. 235 { 240 . INSTICC, SciTePress ( 2016 ).

10. Bureau of Public Roads. Tra c Assignment Manual . US Department of Commerce, Washington DC ( 1964 )

11. Patriksson , P: The Tra c Assignment Problem: Models and Methods . Utrecht: VSP ( 1994 ).

12. Wenzel, S. , & Peter , T. Comparing di erent distance metrics for calculating distances in urban areas with a supply chain simulation tool . Simulation in Produktion und Logistik , pp. 119 { 128 ( 2017 )