Publieke eindrapportage synchromodalit 2016 by TKI Dinalog

SynchromodalIT

EINDRAPPORTAGE R&D PROJECT

IT Services for Synchromodality

SAMENVATTING Three major challenges have motivated the SynchromodalIT project: the need for a unified intermodal logistic network, the need to increase the efficiency and sustainability of logistic services, and the ambition to trigger and push the “mental shift” of both shippers and 4PLs, towards a synchromodal way of working, in which the decision about the mode of transport for the next part of a route is made as late as possible, allowing optimal flexibility in routing and improved quality of service and sustainability. Therefore, the main objective of this project was to enable efficient, reliable, and sustainable delivery of logistic services and strengthen the Dutch logistic sector through: • the development of related planning and scheduling policies and algorithms, and of decision support through serious gaming. • the development of data mining and analytics techniques for large volumes of logistic open source data. • the design of an synchromodal logistic data model, and of an integrated service platform. The SynchromodalIT project has been structured in three research lines (each corresponding to a work package and a PhD project) and two work packages focusing on the dissemination and valorisation aspects. The project is linked primarily to two of Dinalog’s innovation themes: Synchromodal Transport and Cross Chain Control Centers. New planning algorithms, architectures and information services for efficient and sustainable synchromodal logistic service provisioning formed the core of this project. The Cross Chain Control Center concept is embodied in the new SynchromodalIT integration platform, which may serve as real-time process coordination, monitoring, and execution engine. Furthermore, through SynchromodalIT’s focus on serious gaming, the role of human planners in executing logistical innovations and in the optimization of logistic chains is given a lot of attention. This aspect is particularly relevant in a multiple stakeholder setting, such as synchromodal transport, where time sensitive decision-making processes with high levels of complexity and uncertainly are required, and assume the interaction between systems and human planners. Preliminary experiments with the two synchromodal serious games developed during the project show that serious gaming stimulates analytic and strategic thinking, and improves decision making performance under time pressure. In our view synchromodality is no longer a matter of choice: it will very soon become a necessity in order to increase the efficient utilization of the current multimodal logistic infrastructure. This means that, smart, real-time and data-driven synchromodal planning will be given a new sense of urgency from the economic and sustainability points of view. We argue that continuing this research with new approaches on anticipatory and dynamic planning that include new forms of uncertainty and complex emergent behaviors detected in big data generated by logistic cyber-physical systems of systems is especially needed. Furthermore, the incorporation of such approaches in the design of hybrid sociotechnical decision support systems based on the concept of intelligence amplification will lead to a novel type decision support artifact that will enable the partial automation of decision making based on various decision patterns learned from the humans.

INHOUDSOPGAVE

AANLEIDING...........................................1 UITDAGING..............................................................................1 PROJECTOPZET.....................................................................3 RESULTATEN...........................................................................7 ERVARINGEN........................................................................12 TOEKOMSTVISIE...................................................................14 PROJECT PARTNERS..........................................................16

Nothing is quite as challenging as a practical problem. SynchromodalIT is yet another proof that the business and technical complexity of every day logistics decision making is the toughest reality check for any scientific approach in the field. Maria Eugenia Iacob

AANLEIDING In recent years, the logistics industry remarkably changed in that the planning and monitoring of logistics functions is no longer a task performed by customers of logistics service providers (LSPs) (e.g. vendors, manufacturers), and often not even by LSPs, but by a number of so called value-added logistics or fourth party logistics service providers (4PLs, term coined by Accenture). 4PLs provide basic logistic services such as transportation, handling, storage of goods, packaging, clearing of goods, etc., but also take care of the coordination of all LSPs involved, and have the responsibility for the overall process and its quality of service. Due to their supply chain orchestrators role, there has been a growing interest among 4PLs in intermodal transportation. Intermodal transportation involves transport of freight using multiple modes of transport and is attractive because it represents the promise of a more efficient and sustainable logistic process. A recent development in this sense is the study of synchromodal transportation, which concerns an intermodal transportation network where the choice of modality is not fixed up front, but is made for each order individually and might even change during its execution. Thus, the synchromodality concept is synonymous with the creation of an optimal, flexible, efficient, and sustainable transportation in which 4PLs can choose not just among logistic providers but also from a range of modalities at any given moment (through easily switching between modalities if necessary), and for any given order. It is therefore easy to see why the planning process becomes even more critical and challenging in the context of synchromodality. Depending on specific requirements, the synchromodal planning process includes (i) selecting the best combination of modalities and corresponding carriers for each order, (ii) defining the service chain and detail planning of each order, (iii) monitoring the logistic processes and dealing with disruptions and unexpected events, (iv) managing the contracts with preferred carriers, (v) consolidate LTL shipments or even FTL shipments for barge and rail transport, (vi) managing the division of costs and gains for players involved, and (vii) building long-term forecasts in order to assure a viable and robust logistics process. This requires advanced information systems capable of planning under uncertainty and using not only order information but also (big) data generated in the infrastructure network (e.g., builtin sensors, automated monitoring systems, and various (traffic) information services) in order to achieve a true sychromodal way of working.

UITDAGING

Three major challenges have motivated this project: the need for a unified intermodal logistic network, the need to increase the efficiency and sustainability of logistic services, and the ambition to trigger and push the “mental shift” of both shippers and 4PLs , towards a synchromodal way of working, in which the decision about the mode of transport for the next part of a route is made as late as possible, allowing optimal flexibility in routing, improved quality of service, and sustainability. Therefore, the main objective of this project is to enable efficient, reliable, and sustainable delivery of logistic services and strengthen the Dutch logistic sector through • the design of an synchromodal logistic data model and integrated IT service platform, • the development of related planning and scheduling policies and algorithms, and of decision support through serious gaming. More concretely, the SynchromodalIT’s research contribution is focusing on: 1. The substantial development of scientific knowledge at the interface of, operations research, serious gaming and IT in synchromodal logistics. 2. Enabling data interoperability and to trigger and stimulate the shift from logistic corridors thinking towards a unified logistic network by understanding the information requirements with respect to synchromodal logistics, by investigating the existing data models for the different logistic infrastructure networks, and by integrating them in a unified multi-modal logistic infrastructure data model. 3. Enabling effective synchromodal services based dynamic (real-time) information that reflects the current state of the logistic network. 4. Developing techniques able to predict the future state of the logistic network, that can be applied to big data collections. 5. Facilitating the on-demand automated synchromodal planning, such that, (i) modalities can be selected short before or even during transport, and (ii) a modal mix is selected such that costs and CO2 emissions are minimized, and the service level is preserved 6. Triggering a “mental shift” of 4PLs and shippers towards synchromodality and bring “the human in the loop” by using serious gaming and simulation in combination with intelligence amplification and optimal planning techniques during decision making scenario’s. 7. Designing and prototyping a platform architecture to support synchromodality, that (i) provides solutions for the integration with back-office systems, on-board devices, and other data sources, (ii) provides pluggable information services, business intelligence, monitoring services, and decision support services (e.g., real-time planning adjustments), and (iii) handles interoperability problems in the logistic network (where such problems depend on the partners’ diversity and autonomy, and on the heterogeneity of employed technology/data sources).

PROJECTOPZET The SynchromodalIT project has been structured in three research lines (each corresponding to a work package and a PhD project) and two work packages focusing on the dissemination and valorisation aspects. The project is linked primarily liked to two of Dinalogâ&#x20AC;&#x2122;s innovation themes: Synchromodal Transport and Cross Chain Control Centers. New planning algorithms, architectures and information services for efficient and sustainable synchromodal logistic service provisioning form the core of this project. The Cross Chain Control Center concept is embodied in this project by the new the development of a new integration platform, which may serve as realtime process coordination, monitoring, and execution engine. Furthermore, of particular importance in the project (through the focus on serious gaming) is the role of human planners in executing logistical innovations and in the optimization of logistic chains. This aspect is particularly relevant in a multiple stakeholder setting, such as synchromodal transport, where time sensitive decisionmaking processes with high levels of complexity and uncertainly are required and assume the interaction between systems and human planners. Serious gaming stimulates analytic and strategic thinking.

Real-time & Big Data Goal: This work package is focusing on logistic data management, from both the static and dynamic points of view. The work breakdown has resulted in the following research problems: 1. identification of the information needs and design of a data model and data services suitable for supporting advanced synchromodal planning and monitoring services. 2. real-time data processing techniques to extract information, consolidate the information and explicate a quality or confidence level. 3. develop techniques that use historical data to make short/long term predictions about the state of the logistic infrastructure. 4. data mining tools for extracting static geo-spatial information from historical data. Approach and results: We approached the problems mentioned above from two angles: - Problems 1 and 2 have been tackled together by designing a common data model (CDM) and architecture for the automated retrieval of real-time tracking data that includes planned and actual information about orders, statuses and disruptions to increase the ability of synchromodal planning of shipments. A common data model is a standardized data model definition for a particular application domain that fosters the transfer of data between data sources and back-end systems. Since in our case each tracking data source and back-end system (e.g., planning systems) has its own data format, protocol and encoding, retrieving and using tracking data would be impossible without a common data model. Thus, this approach also provides a solution for the integration and data interoperability problem. The proposed CDM has been designed via a bottom-up approach using results of interviews, observations at different logistics service providers, analyses of open data on web-sites, and serves the information needs of the business processes involving such data. The model is also validated against different industry standards. Based on the proposed architecture and the CDM, a prototype was built that has been tested in real operating conditions with a fourth party logistics company. This prototype is subject of a valorisation project, as it is currently transferred to the market in commercial solutions by one of our partners. 3

- Problems 3 and 4 have also been tackled together. We started from the observation that effective barge scheduling in the logistic domain requires advanced information on the availability of the port terminals and the maritime traffic in their vicinity. To enable a long term prediction of vessel arrival times, we investigated the methods to use the publicly available Automatic Identification System (AIS) data to identify maritime patterns and transform them into a directed graph that can be used to estimate the potential trajectories and destination points. To tackle this problem we used a genetic algorithm (GA) to cluster vessel position data. Then we explained how to enhance the process to allow fast computation of incremental data coming from the sensors, including the importance of adding a quad tree structure for data preprocessing. Focusing on a real case implementation, characterized by partially incomplete and noisy AIS data, we demonstrated how the algorithm can handle routes intersecting the regions with missing data and the repercussions on the route graph. Finally, post processing is explained that handles graph pruning and filtering. We validated the results produced by the GA by comparing resulting patterns with known inland water routes for two Dutch provinces, followed by the simulation using synthetic data to highlight the strengths and weaknesses of our approach.

Planning & Gaming Goals: This work package is focusing on two major aspects: the development of efficient scheduling methods for synchromodal logistics and the validation of these decision support instruments through serious gaming. Serious games are not only means to simulate different real life planning scenarioâ&#x20AC;&#x2122;s but also to allow logistic planners to achieve a mental shift necessary for understanding and adopting the new synchromodal way of working. More specifically some of the most important goals in this work package were: 1. Advanced synchromodal planning algorithms to support offline as well as online and (nearly) real-time decision making. 2. Quantitative models that characterizes the relations between shipping patterns, logistics costs, logistic service levels, greenhouse gas emissions, and commercial benefits of LSPs. 3. Forecasting methods for transportation requests and shipment prices, to be used for decisions regarding consolidation (should transport be postponed in anticipation of a consolidation order) and negotiations with shippers. 4 Discrete event simulation models to evaluate and improve all the above planning algorithms, pricing mechanisms, business rules, and forecasting methods. 5. Serious games that use all the above planning algorithms and pricing mechanisms, business rules, and forecasting methods, in realistic decision making scenarioâ&#x20AC;&#x2122;s in which human planners in interaction with the system can choose from synchromodal planning alternatives and see the immediate effects of their choices. Approach and results: With respect to the first three goals, we considered the planning problem faced by Logistic Service Providers (LSPs) transporting freights periodically, using long-haul round-trips. In each round-trip, freights are delivered and picked up at different locations within one region. Freights have time-windows and become known gradually over time. Using probabilistic knowledge about future freights, the LSPâ&#x20AC;&#x2122;s objective is to minimize costs over a multi-period horizon. We proposed a lookahead planning method using Approximate Dynamic Programming. Experiments show that our approach 4

reduces costs up to 25.5% compared to a single-period optimization approach. We provided managerial insights for several intermodal long-haul round-trips settings and provided directions for further research. This approach has been inbedded in a serious game called “Trucks and Barges”, used for both simulation, validation, and knowledge transfer purposes. With respect to the last two goals, we developed a new concept, called intelligence amplification that aims at improving tactical and operational synchromodal decision-making through agent-based modelling and simulation. This concept has been embedded in a serious game called Synchromania ++. The impact of a particular decision can be tested through repeated simulation runs, and different process parameters can be adjusted until the desired behavior is achieved. This is especially critical in the synchromodal planning scenario that assumes (real-time) planning decisions. The planner’s task is to manage shipments by locally optimizing multiple criteria (costs, service levels, CO2 emissions) subject to many constraints, and competing/conflicting objectives of individual autonomous parties involved in the logistic process. We created in the game environment a synergetic relationship between the human planner(s) and supporting smart agents, endowed with rule configuration and machine-learning capabilities to efficiently organize the decision-making. The main benefit is that agents can process large volumes of routine tasks, while humans will only have to handle tasks requiring experience and creativity. Our first version of the IA framework has been recently published. It involved the analysis and decomposition of the planning tasks. Tasks are classified as either creative (and therefore suitable for humans only), or computational (that an agent could handle). The agent is trained to cover specific tasks, is given some degree of autonomy, and it encapsulates learning mechanisms. The human is given the master role, taking the strategic decisions that require creative thinking, and overseeing (and, if needed, overruling) the decisions of the artificial agents. Thus, the decision making process changes from the planner handling all orders manually to the planner deciding when to invoke a specific agent and whether the agent should run continuously, or on call.

Architecture & Services Goal: This work package is focusing on two major aspects: • the design of a 4C type of IT integration architecture for cross-chain information sharing between relevant logistic business network partners. • the development and prototyping of a service platform for decision and planning support services for, e.g., synchromodal (nearly) real-time planning, dynamic pricing, performance/service levels monintoring, logistic infrastructure information services, etc. Approach and results: A currently unsolved problem for small and medium sized companies is how to change their business processes and IT architecture to improve the smartness of their services. We developed a solution that decomposes the concept of smart logistics in a hierarchy of increasingly advanced levels of adaptability. We also designed an enterprise architecture that contains the elements needed for these hierarchical levels. We showed how this enterprise architecture allows a gap analysis and an identification of the steps needed for logistics providers to achieve a higher level of adaptability in their services. We implemented this architecture in a prototype and present the results of a first test in a focus group of logistics providers.

Valorization 5

Goal: In this WP we focused on the valorization of findings from WP 1, 2 and 3. The objective of the SynchromodalIT project has been to further professionalise the logistics sector. This was facilitated by transferring the results of the project in working solutions at participating companies. Approach and results: Our PhD and MSc students have been involved not only in the development of solutions and tools, but also in the implementation of these solutions in practice at the participating companies. Testing and obtaining feedback on solutions and tools was essential for further improving the quality of solutions and tools, and it resulted in several working prototypes tested in real life settings.

Knowledge dissemination

Naam werkpakket

Goal: The focus of this work package was to inform the industrial and scientific communities, and the projectâ&#x20AC;&#x2122;s direct stakeholders about the projectâ&#x20AC;&#x2122;s results. Approach and results: We used different dissemination channels: we published and communicated our work through scientific journals and conferences, we contributed to several books, we organized several workshops, seminars and internal project meetings, we participated in several Dinalog events, we developed several course materials, we invited our partners to give guest lectures and become directly invoved in the teaching process and student supervision, we developed a massive online course on the topic of Supply Chain Innovation, and we organized several gaming sessions with companies from the logistic sector. Omschrijving werkpakket met doel, aanpak, beoogde resultaten

RESULTATEN We can say with confidence that SynchromodalIT has achieved its original goals. As it wil be seen in the remainder of this report, the project delivered more results that we originally expected: numerous scientific publications, two Synchromodal serious games, several new efficient sychromodal planning methods, a conceptualy new Intelligence Amplification framework based on a multi agent architecture that supports a symbiotic man-machine decision making process, an impressive number of master an bachelor gradution theses, several prototypes and case studies, and valorization projects. The research work will continue for another year, time in which the three PhD theses generated by this project will be finalized and defended. Also, it is to be expected, that in the coming period, more publications will be produced based on the research carried out in this last year of the project, which has been very prolific. Our vision for the future of synchromodality research is that concepts, such as those proposed in SynchromodalIT will lead to an increased utilization of the multimodal logistic infrastructure, and to important performance and sustainability gains. Our preliminary results indicate that some of the proposed algorithms could lead 25% cost savings and important diminuation of empty containers moved. Aditionally, the first experiments we carried out using intelligence amplication in logistic planning under time-presure indicate that the approach is improving substantially the performance of the human planners. We are therefore encouraged to further develop this concept and incorporate machine learning and sophisticated planning algorithms in the design of software agents. MAATSCHAPPELIJKE RESULTATEN CO2 reductie

Kostenbesparing

Between 11% and 25.5% aantal

Vermeden vervoerskilometers Reduction in empty containers moved varying between 38 to 47%. Utilisation rate increase on transport legs

Between 38% and 47% Between 9% and 12%

SECTOR RESULTATEN GecreĂŤerde toegevoegde waarde GecreĂŤerde duurzame arbeidsplaatsen Bereikte bedrijven

waarde

Bereikte MKB bedrijven

Onderzoekers/ studenten nu werkzaam bij bedrijven

aantal 14

WETENSCHAPPELIJKE OUTPUT Master thesis

26 master and bachelor theses

PhD promoties

Wetenschappelijke publicaties

Citations wetenschappelijke publicaties Wetenschappelijke seminars, workshops, presentaties etc.

aantal 27 workshops, seminars and work meetings 50 presentatio ns

RESULTATEN WAAR HET PROJECT TROTS OP IS:

1 2 3 4 5 6 7 8 9 10

The concept of Intelligence Amplification as symbiotic socio-technical framework for decision making in synchromodal transport. The maritime route detection technique in incomplete AIS big data, using genetic algorithms. The Synchromania++ serious game that incorporates the Intelligence amplication framework through multi agent simulation and serves as a mental shift vehicle, measurement instrument for decision making performance, and validation tesbed for the framework. Our planning method using Approximate Dynamic Programming, that reduces costs up to 25.5% compared to a single-period optimization approach. The Trucks and Barges computer game, that is the result of a strong cross work package collaboration within the project. The SynchromodalIT service platform architecture and prototype. The synchromodal track and trace architecture and common data model based on real time data, which will be incorporated in Cape Groepâ&#x20AC;&#x2122;s portfolio of logistic solutions. The method and case study for en-route rescheduling of home deliveries, which will be integrated as decision support system in Simacanâ&#x20AC;&#x2122;s Delivery Control Tower. The heuristic to support real-time consolidation and equipment repositioning decisions in a synchromodal environment. A new spatial partitioning method (and case study) based on Geohash that can help scaling up publish-subscribe messaging systems for processing big spatial data (i.e., road traffic data). 9

Intelligence Amplification and Snchromania ++ Our aim was to tackle the core challenge of scheduling container transport by applying symbiotic decision makers based on the concept of intelligence amplification. By creating a symbiotic bond between a logistics planner and a supporting system, we can organize the scheduling process such that the human gets the role of a manager that defines goals/strategies, supervising intelligent agents that carry out routine tasks. The supporting system consists of intelligent agents, which can process large volumes of data quickly, and cannot be fatigued as opposed to a human planner. The humans deal with uncertainties and other loosely structured tasks requiring experience from different fields. We used a serious game to implement the concept and simulate typical weekly planning tasks of the planners. After incorporating intelligent agents into the game, we evaluated the performance of each entity separately, as well as the score when the symbiotic decision maker is used. The benefit of symbiotic approach is it always performs within the range of the best human player. This becomes even more evident when looking at the lower performing participants, whose results dramatically improve, while for the best player, there is just an slight improvement. We conclude from these results that IA benefits every player, yet the improvement is greater for the less experienced players in comparison to the more skilled ones.

Screenshot Synchromania++

Maritime Pattern Extraction and Route Reconstruction from incomplete AIS data

Route extraction in the vicinity of the Port of Rotterdam

For Dutch LSPs one of the most important resource are barges that ideally are fully utilized with no or minimal waiting times. This rarely happens due to unscheduled arrival deep sea vessels at the port terminal, which are handled with priority. The occurrence of such an event forces barges into a waiting state, until the terminal becomes available to service them again. Therefore, LSPs need an effective way to estimate deep sea vessel destinations and arrival times, to be able to create an optimal scheduling for their own barges. The problem of estimating destination and arrival time for more than one hour ahead requires identification of sailing patterns for all vessels in the region of interest. We developed a novel approach on how to adapt a genetic algorithm to handle the real life data mining problem of pattern extraction. While we use imperfect publicly available maritime AIS data to prove the approach, the same method could be applied to other modes of transport without restrictions. Finally, as maritime routes have the tendency to change over time due to external factors such as weather and tides, the ability of the genetic algorithm to gradually evolve with the new data, will make it the best candidate for providing real time naval pattern awareness for typical logistics enterprises.

ERVARINGEN Âź A4 max. Hoe is de projectuitvoering verlopen?

Open innovatie During this project we learned a lot from our partners. In particular we are very grateful for our technology partners, OV Software Groep, Cape Groep, NexusZ and MX ystems from which we received a lot of support in terms of software solutions, and domain expertise. Also, we must emphasize the critical contribution of our logistic partners CTT, Seacon and Wayz that helped us enormously by providing us with practical problems that led to the identification and definition of the major research directions. They also provided data, and took part in the evaluation of our solutions. Finaly, we would like to mention the invaluable contribution in terms of knowledge and know-how coming from TNO and TXchange that both had a critical role in the development of our two Synchromodal serious games. We believe, that every single partner covered yet another expertise area that was needed throughout the research. In terms of the open innovation process, our experience is that working in small pilot projects in which two or maximum three partners with complementary expertise areas are involved works best. Partners from the same expertise area/sector are generally reluctant to fully disclose details about their own r&d.

Dinalog en Topsector Logistiek Contentwise, Dinalog has contributed in many ways to this project: - It had a very constructive role during the project review discussions, by suggesting us to seek synergy and collaboration with TNO, concerning synchromodal gaming. As result we approached TNO, which eventually became project partner, and joined forces with us to create one of our most successful results: the Synchromania ++ game. - Secondly, some of our researchers have been active in other Dinalog projects (e.g., CargoHitching, CATeLOG, etc.) as well, and transferred ideas and knowledge from and to the project. Due to the active role we played and to the close relationship we developed with Dinalog, our researchers have also actively contributed to the definition research agendas for the Topsector Logistics. - Finally, through its congresses and events Dinalog offered us excellent opportunities to promote our results, and reach a broad audience and many companies that are active in logistics innovation and research.

An architecture and Common Data Model for Open Data-based Synchromodal Cargo-Tracking

Wouter Bol Rap, Cape Groep

In logistics, questions as “Where is my container?” and “When does my con-tainer arrive?” can often not be answered with sufficient precision, which restricts the ability of logistics service providers to be efficient. Since logistics is complex and often involves multiple transportation modes and carriers, improving efficiency and saving costs in the supply chain requires communication between the different parties and the usage of real-time data is critical. Currently, LSPs use real-time data to a very limited extent, mainly for tracking the progress of a specific part of a given shipment. This data is retrieved manually from a number of websites and sharing with other actors is not even considered. This leads to lack of end-to-end visibility and delays in planning. This master thesis research proposes an architecture and a common data model (CDM) for an integration platform that allows the automated collection of real time container tracking data enabling LSPs to plan more efficient. Currently, there is no CDM available that contains all the information required and enables LSPs to track their shipments real-time. The CDM we designed includes planned and actual information about orders, statuses and disruptions to increase the ability of synchromodal planning of shipments. We took a bottom-up approach using results of interviews, observations at different logistics service providers, analyses of open data on websites, and serves the information needs of the business processes involving such data. The model is also validated against industry standards. Based on the proposed architecture a prototype was built that and tested in real operating conditions with a 4PL. Some preliminary estimates indicate that it can save 700 working hours on a yearly basis while increasing the ability to execute synchromodal shifts since planners are able to make a more efficient planning based on the more accurate data the platform retrieves.The plan is to incorporate this prototype in Cape Groep’s commercial solutions.

En-route rescheduling of home deliveries Simacan’s Delivery Control Tower (DCT) is a software product designed for monitoring and communication in home delivery operations of retailers. Since the recent introduction of this product at a large Dutch retailer, a team of six people uses the product to monitor the home delivery operation, communicating with the vehicle drivers as well. Real-time vehicle positions and expected arrival times are calculated for all customers on delivery routes in the DCT. In case of expected lateness of delivery, drivers call the control room. They determine en-route rescheduling solutions to reduce the number of late deliveries. The current rescheduling process is labor-intensive, and non-optimal rescheduling solutions are often proposed. Therefore, Simacan came with the assignment of developing an en-route rescheduling method that automatically triggers on lateness, suggesting the best rescheduling solution. This method had to be implemented as a Decision Support System (DSS) for control room teams to receive automated suggestions for en-route rescheduling. Our solution was to solve the en-route rescheduling problems using a Tabu Search Rescheduling (TSR) method. To test the TSR method performance, we also developed a simulation model. Our TSR method is able to reduce the number of late deliveries with 63% for new customers and 24% for recurring customers in our simulation model. No more manual work by the control room is involved because we automatically trigger and calculate rescheduling solutions.

Peter Bijl, Simacan

TOEKOMSTVISIE Synchromodality is no longer a matter of choice: very soon it will become a necessity. It is obvious the road infrastructure is reaching globally its physical capacity limits, and it has become one of the major polluters, and global warming factors. Building more roads would only be a partial and temporary solution for the capacity problem, and would worsen the environmental one. Therefore, until a completely new transport paradigm and infrastructure will be invented and built we will have to get things done with the infrastructure we have now, by using it in a more efficient way. That means that we will have to exploit the existing free capacity in the water and rail networks, which are both cleaner, and cheaper. This is why we think that efficient, smart and real-time data-driven synchromodal planning will become increasingly important from the economic and sustainability points of view. Investing in this type of research will place The Netherlands in the pole-position. Following this line of reasoning, we define our vision for the future of synchromodality along the following lines: In our research, we showed that including demand uncertainty in the scheduling of synchromodal freight transport pays off when considering networkwise and multi-period performance. The gain was due to a change in the scheduling perspective to more anticipatory and dynamic decisions such as postponing transport of an order for an expected better consolidation in the future, or such as transporting a freight earlier to a farther away terminal to use free capacity in a highcapacity mode. Furthermore, we observed that the improvements in performance of anticipatory and dynamic scheduling was dependent on some synchromodal characteristics more than on others (e.g., it is better to have pre-announced orders than a balanced network). These observations raised further questions such as: 1. How can other new forms of uncertainty be included in the scheduling of synchromodal freight transport and what is their expected influence on performance? New forms of uncertainty are those parameters assumed to be fixed before, but now available as probabilistic information due to the advances in IT measurements. We should investigate if they can be included in the scheduling, and what the gains are of doing so. 2. How can the planning processes for the crucial synchromodal characteristics (e.g., pre-announced orders, un-balanced destinations) be included in the scheduling of synchromodal freight transport? The focus of our search has been on the transport processes of the carriers and LSPs, but including manufacturing and distribution scheduling could have a larger impact on the network-wise multi-period performance. Such “full supply-chain scheduling” research has been done before (e.g., the inventory routing problem), but on a unilateral organization perspective. With the trend of the sharing economy for individuals, this is an intriguing question for synchromodality as well, especially from a planning perspective. Our research into intelligence amplification also showed additional benefits and possibilities. We observed positive improvement in human’s behavior in the form of adapting to the new situation and developing new and better strategies to handle tasks in the logistics. Also, by incorporating ongoing research into the way for the agents to learn and adapt to the human, it will be possible to collect and analyze decision patterns, strategies, goals, and priorities. With the ability of agents to rapidly share large amount of data we hope that in the future it will be possible to come up with a form of a novel decision support artifact that will allow collective decision making – i.e. utilizing various decision patterns learned from the humans to bring awareness to the possibilities that could not be identified by the humans or machines on their own. Finally, in terms of information technology we believe that synchromodal logistics will significantly benefit from paradigms such as the Internet of Things, Cyber-physical systems of systems, and big data analytics. The rapid development of these new technologies increase the criticality of designing performant supply 15

chain integration solutions, and of pluggable and cloud-based information service platforms, in order to reap the benefits of this distributed, rich and heterogeneous data infrastructure.

Vervolgactiviteiten Looking back at the lessons we have learned, we can say that one type of partner we have missed in the consortium was the rail operator, mostly because we only had very limited access to data for this modality. Also, because of the considerable size of the consortium, the management of the project was challenging at times. It would have been probably better to allocate resources for a professional project manager. From the research and innovation point of view, our experience is that working in small pilot projects in which two or maximum three partners (with complementing expertise areas) are involved works best.

â&#x20AC;&#x153;Synchromodality is no longer a matter of choice: very soon it will become a necessity.â&#x20AC;?

PROJECT PARTNERS SynchromodalIT has enormously benefited from a stable and comprehensive consortium. The knowledge transfer has been bidirectional: the knowledge partners have absorbed a lot of domain knowledge, data and concrete business cases/problems from the industry, and research results have been transferred to industry through implementations, validation projects, and through our master/PhD students. As seen from profile and distribution of partners, our consortium is • Comprehensive: it ensures a full and balanced coverage of all logistic chains, and of the supporting research disciplines, and technologies: 4PLs and 4PL solutions (Seacon, CTT, Wayz), IT support and system integration (Cape Groep, OVSoftware, Simacan, RGB+, NexusZ, MX Systems, Arcadis), research and innovation (UTwente, TNO, TXchange Thales). • Robust and stable: The SynchromodalIT consortium has been built upon outstanding IT-driven logistics innovation track-records, and long lasting past business, and research collaborations, and partnerships. This is why no partner has left the project and we managed to attract new partners: TNO and Seacon. • Highly multidisciplinary: not only our consortium consists of essentially different types of companies (IT, R&D, consulting, technology, logistics), but also the experts coordinating the research have different disciplinary backgrounds, complementing each other: IT, information systems architecture, supply-chain management, operations research, serious gaming. Furthermore, all of them are active in the field of logistics innovation, as proven by their numerous high quality publications and projects.

PRIVATE PARTNERS OV Software Groep Organisatie: OVSoftware Groep consists of several companies having each its own IT focus. OVSoftware is developing software using Java en .NET technology for companies and government organisations in the regions Arnhem, Amersfoort, Apeldoorn en Den Haag, east of The Netherlands, and Munster (Nordrheinwestfalen, Germany). Next to OVSoftware, the Group also consists of RGB plus automatisering b.v. in Raalte, Simacan in Amersfoort, and Docdynamic in Oldenzaal. The OV software Groep and its companies have become one of the core partners in the consortium in terms of the size and importance of its contribution: they hosted many of our master students (from three different study programs) and assigned them research topics of direct relevance for our project in all three research lines defined in the three work packages, they contributed directly to the delivery of many results and incorporated the obtained results in their software solutions thus contributing to the valorization and knowledge transfer, and they also hired some of our students.

Simacan and RGB+ Simacan is a young Dutch company founded in 2013. Simacan products integrate real-time traffic information in primary business processes, and are focusing on making geospatial data truly accessible and useful. More precisely, Simacan offers primarily four types of IT solutions: The Simacan Control Tower facilitates shippers and transporters in keeping overview and control of their planning and operation; the Simacan Traffic Center that enhances the operational overview and daily efficiency in traffic control centers and operations rooms, the Simacan Traffic Cast that speeds up the publication of relevant traffic information 17

for service providers and media organizations , and the Simacan Developer allowing the integration of real-time and historical data sources in client services via powerful web APIs.RGB+ is a dynamic innovative software development company with expertise in the area of software solutions for the transport and logistics sector. In particular they offer cloud solutions for Transport Management Systems, Fleet management and real-time Track and trace of temperaturecontrolled shipments.

Cape Groep Organisatie: Cape Groep is a system integrator and software development company. They provide model-driven agile solutions primarily to the Dutch logistics industry in regard with agile software development, connectivity, integrations, supply chain control, business intelligence, integrated mobile apps, reporting and cloud computing. Cape Groep is in particular active in the areas of data integration and control tower solutions, and activity-based real-time carbon management systems, which are key to synchro-modal thinking, planning and management. They also offer real-time track & trace solutions based on open source data. Some of Cape Groepâ&#x20AC;&#x2122;s important customers are PostNL, Seacon logistics, Kuenhe + Nagel, Mammoet, Heisterkamp, etc. Cape Groep, assumed a critical a multifacetted role in SynchromodalIT: they provided the project with the development and integration technology that was critical for the prototyping work in the third workpackage, and incorporated the obtained results in their software solutions thus contributing to the valorization and knowledge transfer, they have been one of the most active valorization partners by translating many concepts we developed in working prototypes, they have participated in the development of teaching materials and activities at the UT and Saxion, provided validation and teachin cases cases, hosted and hired some of the SynchromodalIT researchers (both students and PhDs).

NexusZ NexusZ.com is a company that builds decision support software that: provides (geographical-) location-based intelligence services, integrates complex, often real-time, data-streams and other information sources, and supports users in their decision making process. NexusZ delivers customized solutions for the following sectors: logistics, (sustainable) energy, healthcare, safety & security, and for market research. The collaboration with NexuxZ has resulted into one publication and transfer of sailing routes knowledge into the project. Also, NexusZ provided the project with the data infrastructure necessary for the collection of AIS data. This real-time data was critical for the maritime routing pattern discovery research.

MxSystems MX Systems has extensive experience with large and complex IT projects in areas such as water data management, hydrological solutions and shipping corridor management systems. For the port of Rotterdam MX Systems is developing a complex system for the simulation and prediction of water levels and stream speed. MX Systems is also developing an information system for the vessel traffic service in the Schelde region, which is part of a complex integration architecture with other chain partners. MX Systems hosted several master students that worked on

problems related to water management based on real-time data.

Arcadis Arcadis is the leading global Design & Consultancy firm for natural and built assets. Arcadis is offering deep market sector insights and collective design, consultancy, engineering, project and management services. Arcadis has 27,000 people active in over 70 countries that generate €3.3 billion in revenues. In the area of logistics, Arcadis has extensive experience and wide-ranging expertise to plan, design, build and manage the logistics infrastructure of the future. This is complemented by their Information management department, through complex GIS software solutions for the real-time geographical mapping and visualization of data. This has been also made available for our project, for the visualization of maritime routes of large deep see container vessels.

TXchange - Thales TXchange is a research collaboration between Thales Research & Technology and University of Twente – which researches, designs and develops serious game artefacts. Over the last 8 years TXchange did research on game based learning and delivered various serious game systems (board and digital games) in many application fields. Its current focus is on adaptive 2D narrative game engines and games. In addition, TXchange provides technology for elucidating behavioural models through game analytics. In SynchromodalIT we have used the TXchange game design approach for the development of our “Trucks and Barges” game. Also some of the TXchange games have served as inspiration for our gaming scenarios.

Wayz Wayz researches/surveys logistics flows, advises on optimization of logistic processes, and supports shippers and logistic service providers in implementing innovative and sustainable projects. Wayz collaborates with knowledge institutes and government agencies on innovative logistics projects. Wayz’s core themes in this area are Control Towers, Synchromodality, and Sustainability. Wayz made available for our researchers its know-how in synchromodal logistics and helped us identified some of the core research problems that stem from the logistics practice, such as the prediction of arrival times in particular for deep sea vessels, which would allow a more efficient planning of barges in the port of Rotterdam.

SEACON Logistics Seacon Logistics is a large logistics chain orchestrator (a so-called 4PL) with a maritime character. In collaboration with their partners, they offer total logistic solutions based on their worldwide presence in more than 75 countries. Seacon Logistics has been leading for more than 30 years in applying the multimodal transport concept. In terms of IT solutions, Seacon has developed (together with the UT, Cape Groep, and TNO) and is using one of the first operational synchromodal control towers. This control tower 19

is taking care of exchange of information flows throughout every link in the logistics chain, such as customs, shipping companies, transport organisations and banks, and the IT systems of the shippers. It also has advanced business analytics, monitoring, planning, warehouse and transport management capabilities. Seacon has played an active role in the project by hosting several of our master students and providing us with case studies.

CTT Combi Terminal Twente B.V. (CTT) has a State-of-the-art inland terminal in the Netherlands. In cooperation with the Port of Rotterdam it provides container shipments to Twente, the German hinterland and many other European final destinations. With four terminal locations, water-, railand road transport facilities supplemented by a private warehouse, CTT offers all the benefits of synchromodal transport to its customers. CTT is actively involved in IT innovation, and participates in a multi-sided platform for controlled data sharing and intelligent apps, built by entrpreneurs from the Port of Twente community. CTT has played an active role in the project by hosting several master students, and providing us with case studies. CTT also regularly participates in our student hackathons, by challenging them with very interesting practical problems.

PUBLIEKE PARTNERS TNO TNO (www.tno.nl) is a Dutch research institute with over 3000 employees that connects people and knowledge to create innovations that boost the sustainable competitive strength of industry and well-being of society. TNO was founded by law in 1932 to enable business and government to apply knowledge. As an organization regulated by public law, TNO is independent: not part of any government, university or company. The organization also conducts contract research, offers specialist consulting services, and grants licenses for patents and specialist software. TNO tests and certifies products and services, and issues an independent evaluation of quality. Moreover, TNO sets up new companies to market innovations. TNO has extensive knowledge of national and international mobility and logistic systems. This is supported by laboratory and 'real-world' measurements and simulations of relevant systems and subsystems. Examples of this work are: measurements of vehicle emissions, both in the lab and in real traffic; modeling of traffic emissions and development of emission factors; assessment of vehicle and driveline technologies, monitoring of pilots with sustainable technologies; concepts and technologies for environmentally-friendly dynamic traffic management; strategies for The Netherlands as a hub in in worldwide freight chains; concepts and agreements for design, planning and management of common logistic processes; use of IT for optimizing the performance of (international) logistic chains, development of trade-logistic concepts based on data sharing. Role in this project: contributes primarily to WP2, by providing expertise and games in the area of synchromodal serious 20

gaming.

UT – Behavioural, Management and Social Sciences, and Electrical Engineering, Mathemeatics and Computer Science is a Dutch university that offers degree programmes ranging from business administration, psychology to applied physics, engineering and biomedical technology. The University of Twente believes that a multidisciplinary approach will stimulate innovation in science and society alike. Patents, lifelong learning programmes, and spin-off companies testify to this commitment, as does UT’s intensive involvement in research that enhance knowledge infrastructure in the Netherlands. So far, the UT has produced over 1,000 spin-off companies; more than any other Dutch university. The University of Twente also has a world class research programme. Academic Ranking of World Universities (ARWU) ranked the University of Twente 76-100 overall in the world in the field of Engineering. University of Twente also holds the title of the most entrepreneurial university in Netherlands and was voted the best Dutch university in the area of valorization. The UT’s research group (IEBIS, SCS) from which this project stems, have produced high-quality interdisciplinary research and education in area of industrial engineering, business information systems, and pervasive systems. They closely collaborate with industry, knowledge institutes and government agencies. Their relevant research for logistics, concerns decision support systems and inter-organizational systems connecting networks of businesses and governments. Examples of work include novel ways of organizing networks such as dynamic global sourcing and multi-agent coordination, sensor networks, cyber physical systems, quantitative models and algorithmic approaches, simulation and gaming, IT architecture, business modeling, design, planning, optimization and control of operational processes in production, transportation, supply chains, and service logistics. UT’s role in this project was leading. The UT acted as coordinator and project manager, led and carried out most of the research work, and facilitated the interaction between and with the partners. The UT also initiated and contributed to the majority of the dissemination activities.

Naam partner Omschrijving partner: beschrijving organisatie en rol en specifieke inbreng expertise in het project

DUTCH INSTITUTE FOR ADVANCED LOGISTICS Dinalog is het technologisch top instituut van de topsector logistiek ter bevordering van innovatie in de logistieke sector. Publicatiedatum: www.dinalog.nl