Charging infrastructure planning is extremely complex! There are many components and stakeholders that have to work together while focusing on their own and societal objectives. Moreover, the optimality of any charging infrastructure plan heavily depends on the transport energy demand that the charging infrastructure components serve. In particular, a movement trace in relation to a plan defines how energy is consumed and can be acquired along the trace. Thus, movement traces are paramount in evaluating the benefits of a placement and dimensioning of components in a complex system, which then needs to be weighed against the costs.

At Gordian we provide planning tools that are powered by lightning-fast analytics and artificial intelligence. The tools allow to interactively evaluate or optimize a plan such that it best serves (covers) the transport demand (movement traces) with a set of infrastructure components. A non-optimal plan and an optimized plan for charging components (shown with filled circles) are illustrated in the grey and orange images for a single movement trace. Now imagine optimizing the plan for millions of traces. And let’s not forget that there are an astronomical number of possible plans. The speed and information in our analytics allow unprecedented optimizations: we accurately evaluate the effects of millions of the top plans against millions of movement traces under a few minutes. Thereby, we provide evidence-based decision support and up to 5 times more cost-effective plans than our competitors. We provide this for single or multiple stakeholders and objectives.
Watch electric road video ! Try our electric road planning demo tool ! Contact us for more info Close

Transport and transfer hubs, both for people and goods, are most cost-effective if they are located at places that are along many movement traces and are either close to where the traces start, end or merge. At Gordian we believe that the keyword is along because using a hub that is off a given movement trace ultimately represents a detour, which one logically wants to minimize. With all other things considered equal, which hub location with bus and shared mobility services is more attractive: one that the traveler has to walk back 500 meters or one that is 500 meters ahead along the planed and direct path of the traveler?

Similar to the other use cases, at Gordian we provide planning tools that are powered by lightning-fast analytics and artificial intelligence. The tools allow to interactively evaluate or optimize a plan such that is best serves (covers) the transport demand (movement traces) with a set transport hubs. A non-optimal plan and an optimized plan for one hub (shown as a circle) are illustrated in the grey and orange images for four movement traces. Now imagine optimizing the plan for millions of traces. And let’s not forget that there are also an astronomical number of possible placements and dimensioning of hubs. The speed and information in our analytics allow unprecedented optimizations: we accurately evaluate the effects of millions of the top plans against millions of movement traces under a few minutes. Thereby, we provide evidence-based decision support that increases the utility of the planned hubs.
Try to beet Gordian at optimizing transport hubs !
Contact us for more info Close

Transit route networks (e.g., a set of buss lines) are most cost-effective if they are dimensioned and located at places such that the maximal number of travelers can accomplish their trips from start to end with as few and possible transfers and as direct as possible routes. At Gordian we believe that movement traces, in particular, simulated least-costs (shortest / fastest / cheapest) movement traces from origin-destination transport demands reflect the underling transport demand that the transit route networks should serve.

Similar to the other use cases, at Gordian we provide planning tools that are powered by lightning-fast analytics and artificial intelligence. The tools allow to interactively evaluate or optimize a plan such that is best serves (covers) the transport demand (movement traces) with a set of transit routes. The grey and orange images illustrate a non-optimal and an optimal plan with some existing and new routes (shown as lines) that aim to serve the transport demand (shown as circles), which arises from how well the transit route network serves the movement traces. As with the other use cases, there are millions of traces and an astronomical number of possible placements and dimensioning of transit routes. The speed and information in our analytics allow unprecedented optimizations: we accurately evaluate the effects of millions of the top plans against millions of movement traces under a few minutes. Thereby, we provide evidence-based decision support that increases the utility of the planned transit routes.
Contact us for more info Close

The cost of commercial transport today is generally equally divided between the cost of the vehicle, the fuel and the driver. Electrification and automation will radically change these costs. Full automation of heavy transports is challenging in urban areas and building up the communication infrastructure to support safe autonomous driving everywhere is costly. For such an application one needs to find a set of hub-to-hub corridors on which autonomous electrified transports is cost-effective. Apart from from meeting minimum physical requirements, such corridors are cost-effective if the length of the corridors that connect hubs and support automated transports is relatively long compared to the off-corridor transport route parts that need to be human-controlled. Hub-to-hub corridors should ideally form a connected network and reutilize the same infrastructure.

Similar to the other use cases, at Gordian we provide planning tools that are powered by lightning-fast analytics and artificial intelligence. The tools allow to interactively evaluate or optimize a plan such that is best serves (covers) the transport demand (movement traces) with a network of hub-to-hub corridors. The grey image illustrates some traces out of which the transports between north and south could be more cost effectively served by a hub-to-hub corridor (shown in orange on the other image). As with the other use cases, there are millions of traces and an astronomical number of possible placements and dimensioning of hub-to-hub corridor network plans. The speed and information in our analytics allow unprecedented optimizations: we accurately evaluate the effects of millions of the top plans against millions of movement traces under a few minutes. Thereby, we provide evidence-based decision support that increases the utility of the planned hub-to-hub corridors.
Contact us for more info Close

A vehicle that moves with unused cargo space is not cost-effective, but it happens all too often in practice. Even if routes are well planned for a fleet, often deliveries and pickups are served by different routes. On delivery routes cargo space is freed up as goods are delivered, while on pickup routes the vehicles travel with empty cargo space that gradually fills up. The cargo and the unused cargo space along some pickup and delivery routes, or parts of routes, complement each other. At Gordian, borrowed from the charging infrastructure planning use case, we treat cargo space in a similar way as the empty battery capacity of moving vehicles and cargo as the energy in the battery that after a delivery disappears. Using this approach we find delivery and pickup routes that complement each other and thereby fully utilize the moving cargo space.

Similar to the other use cases, at Gordian we provide planning tools that are powered by lightning-fast analytics and artificial intelligence. The tools allow to interactively evaluate or optimize a plan such that transports from complementing routes are consolidated. The grey image illustrates one delivery and one pickup route which can effectively be consolidated into a one pickup-delivery route shown in orange on the other image. As with the other use cases, there are millions of traces and an astronomical number of possible consolidation plans that have to be extracted and scored from millions of logistics routes. The speed and information in our analytics allow unprecedented optimizations: we accurately evaluate the effects of millions of the top plans against millions of movement traces under a few minutes. Thereby, we provide evidence-based decision support that increases the utility of the planned consolidated logistics.
Contact us for more info Close

Traditionally, distributions warehouses are used to merge several shipments from different suppliers to the same customer. Distribution warehouses are static, capacity constrained and expensive. Meanwhile, transport demands in supply chains are dynamic. Merge-in-transit (MIT) can eliminate the need for distribution warehouses if one can discover these dynamic consolidation opportunities and can adjust the operations to exploit them. We at Gordian believe that the answer again lies in the movement traces of shipments and vehicles and their dynamic orchestration in space and time. As a side note, planning the formation of heavy freight convoys is very similar in nature to MIT planning and also has links to electric charging infrastructure operation for electrified transports.

Similar to the other use cases, at Gordian we provide planning tools that are powered by lightning-fast analytics and artificial intelligence. The tools allow to interactively evaluate or optimize a plan such MIT operations. The grey image illustrates thee shipments from three suppliers to the same customer and the orange image illustrates how these are consolidated at the merge point into a single delivery. As with the other use cases, there are millions of traces of shipments and an astronomical number of possible MIT plans that have to be extracted and scored from millions of logistics routes. The speed and information in our analytics allow unprecedented optimizations: we accurately evaluate the effects of millions of the top plans against millions of movement traces under a few minutes. Thereby, we provide evidence-based decision support that help discover and exploit these MIT opportunities.
Contact us for more info Close

Vehicles are most cost-effective when they are intelligently dispatched and repositioned. Hence, intelligent fleet management can be regarded as a crucial task in maximizing revenue and QoS in a transport system. However, optimizing the positioning of fleets is complex and depends on the real-time evolution of demands in time and space as well as the state of the vehicles. At Gordian, we therefore believe that pro-active demand prediction is a key component to optimize fleet management strategies. Techniques from AI, such as deep reinforcement learning (DRL), can be used to determine policies which maximize QoS and revenue metrics while adhering to specific constraints. Large numbers of movement traces are essential to form state spaces which can be interpreted by the AI models and turned into optimized fleet management strategies.

At Gordian we provide planning tools that are powered by lightning-fast analytics and artificial intelligence. The tools allow to optimize fleet management policies with the aim of satisfying future demand, shortening dispatch times and maximizing vehicle utilization. The grey image shows four zones of demand, which can be conveniently derived from millions of movement traces using the tools that are provided by Gordian. Based on the predicted demand and the real-time state of vehicles, the AI model is able to determine the optimal routes for every vehicle such that the future requests can be served efficiently while maximizing the Quality of Service (e.g. low waiting and detour times). Moreover, the tools enable the user to incorporate any kind of objective into the optimization task. Thereby, we provide evidence-based decision support that facilitates pro-active intelligent fleet management.
Contact us for more info Close