Technical Field

[0001] The invention generally relates to a method for transport analysis and to a processing system for transport analysis and more particularly to a data compression method and to a corresponding processing system useful in the field of transports.

Background Art

[0002] In premium logistics, goods are often transported between different countries or even continents. The start and end points of a transport are usually very specific, for instance, the pick-up point may be at the residence of company A and delivery point may be at the residence of company B. This leads to a complex and widespread web of transports. The transport routes may change constantly, due to different customer demands, different restrictions imposed on traffic (traffic jams, roads under construction etc.), availability of transport means and other factors. Thus, the structure is not only complex but also varying over time. This makes efficient planning of premium logistics services extremely difficult.

[0003] For a logistics provider to work effectively, a very important feature is to be able to efficiently make use of co-loading, i.e. to use one transport means for different shipments. This only makes sense, of course, if at least a part of the transport paths for the different shipments is or could be identical. The higher the number of transports and the more complex and widespread they are, the harder it is to find such (potentially) coincident route parts which may be used for co-loading. In the case of co-loading, as well as in other cases, it is important to effectively combine information regarding different transports in order to discover relations. This can only be done by efficiently processing and analysing a great number of transport routes.

Technical Problem

[0004] It is thus an object of the present invention to provide a way of effectively processing data of a transport network. This object is solved by a data compression method according to claim 1 and by a processing system according to claim 14.

General Description of the Invention

[0005] The invention provides a data compression method for compressing transport data by a mapping procedure using node data. It is understood that some kind of processing system is used to perform the inventive method, which processing system can be more or less complex. Usually, it may comprise at least one processing unit and at least one memory device. These components of the processing system could e.g. be constituted by a conventional PC. The transport data and the node data are provided to the processing system. Usually, the data will be stored in a memory device either part of or accessible by the processing system. It should be noted that the present invention is not limited to the transport of goods, but also encompasses the transport of people or livestock.

[0006] The transport data define a plurality of paths on an at least two-dimensional map, each path describing a transport. The map could also be referred to as a coordinate space. The map will be a representation of an area in which the transports to be analysed take place. E.g. it may be a two-dimensional representation of Europe, America, etc. Normally, two dimensions will be sufficient to characterise a path. It is, however, conceivable to include further dimensions. Each path describes a transport, i.e. the route of a transport means like a van, a truck and the like. In a wider sense, a "transport" may also refer a route taken by an empty transport means, e.g. a van that does not transport any goods on this specific run. Normally, the data will be a (two-dimensional) coordinate representation of the path, e.g. corresponding to latitude and longitude.

[0007] The node data define a set of nodes on the map, each node having a proximity zone. Here, again, the data will usually correspond to (two-dimensional) coordinates of each node. The nodes are pre-defined before the actual data compression takes place. In particular, a node may correspond to a significant point for transport, like the residence of an important client of a logistics provider, important crossroads, a city, a cross-dock, a warehouse, a railway station, an airport etc, conveniently placed to cover most of the geographical area the logistics provider operates in. The nodes are defined as a set, i.e. there is no specific "sequence", although each node may have an identifier like a name or a number. Furthermore, each node has a proximity zone. The proximity zone, of course, is a zone surrounding the node. If the map is two-dimensional, the proximity zone is also two-dimensional. The zone can be circular, i.e. it extends to the same distance in any direction around the node. It is, however, conceivable to define a zone of more complex shape, e.g. if there is a geographical obstacle (like a mountain or an unbridged river) on one side of a node, the proximity zone could only extend up to this obstacle. Apart from the shape of the proximity zone, its size may be different for different nodes, e.g. corresponding to the importance attributed to the respective node. Usually, however, all proximity zones are circular and have the same size.

[0008] For each path, the mapping procedure is performed to obtain sequence data defining a node sequence consisting of nodes through whose proximity zones the path passes. This corresponds to "moving" along the path, i.e. checking the path in the direction-of-travel and finding out if the path enters a proximity zone of a node. If not, this node will not be considered for the node sequence. If the path enters the proximity zone of a node (or rather, at least one node, because proximity zones may overlap), the path is considered to be "close" to this node. A sequence consisting of such close nodes is created (normally by the processing system), although not all close nodes may be part of the sequence, depending on optional further criteria. The sequence data are compressed data with respect to the path data. The compression is achieved by the mapping procedure. In almost every case, this is a lossy compression, i.e. the original path data cannot be extracted from the sequence data. This, however, usually not necessary. Anyhow, the amount of data is greatly reduced by the mapping procedure. The sequence data defining the node sequence(s) can be stored as e.g. as a file which uses much less memory than the original path data. Also, some information derived from the node sequence may be stored, additionally to or instead of the sequence itself.

[0009] Apart from the compression aspect, the mapping allows for a general overview of the course of a transport with respect to the nodes, which usually represent logistically relevant points in the map. This is not only a coarsening process in which the path is reduced to a (usually) small number of supporting points. Rather, when in a proximity zone, the path may be "attracted" by the node (although the path is not changed), in that it is associated with this node even though it does not exactly run through it. From a logistics point of view, it may e.g. be worthwhile to consider rerouting the respective transport - or a similar one in the future - to exactly go to or through this node. This may open the possibility of arranging a co-loading with another transport that is also "close" to this node. This is just one example for the applicability of the results of the mapping procedure. Especially in some embodiments, the method may also be considered as an analysis method. The term "analysis" may refer to very different degrees of analysis. In any case, data referring to transports is taken to derive information that goes beyond the "raw" data. As will become apparent below, some embodiments of the method go beyond data compression or even analysis, but always rely on the inventive compression principle.

[0010] The mapping associates the two or more dimensional path with a sequence of nodes. Although the nodes have coordinates of the same dimension as the path, the sequence itself may be considered as a one-dimensional object. E.g. the sequence data could be a sequence of integer numbers, wherein each integer number corresponds to a node. In this regard, the mapping procedure leads to a reduction of dimension, a data compression which saves on computing power and time. The concept of the compression method is to abandon information that is not important from a logistics point of view. This is to some extent comparable to the MP3 compression, where sound information is abandoned that is not important for the listener's perception.

[001 1] Usually, each path is defined by a sequence of waypoints. Consecutive waypoints (or supporting points, in mathematical terms) may have equal distances. It has been found that distances between 5 and 20 km are very useful, with a distance of 10 km being especially preferred. Of course, each waypoint is represented by (usually two) coordinates, normally corresponding to latitude and longitude. For such a sequence of waypoints, the mapping procedure comprises checking for each waypoint whether it is within a proximity zone of at least one node. Obviously, a characterisation by waypoints, whose number may be kept relatively small, allows for a reduction of memory needed for each path and also for a relatively quick completion of the mapping procedure.

[0012] In a case where proximity zones overlap, the mapping procedure could lead to "unstable" or complicated results if a path runs through such an overlap zone. To avoid such problems, it is preferred that if a waypoint is in the proximity zone of several nodes, only the closest node is added to the sequence. The method therefore not only checks whether the path is in a proximity zone of a node, but also if this is the closest node with respect to the part of the path (usually, the waypoint) under consideration. From a logistics point of view, nodes that are further away (although in "proximity") are considered less important.

[0013] The mapping procedure may consider all nodes defined on the respective map. However, it may be wise and timesaving to do some pre-selection. Any reasonable route of a transport will not move outside a certain area comprising the start point and end point. For example, if a path starts in southern France and ends in northern Spain, it does not make any sense to consider nodes located in Germany. Therefore, it is preferred that before each mapping procedure, a subset of regional nodes is selected from the set of nodes and only proximity zones of regional nodes are considered. Of course this could also include the - trivial - case that all nodes are selected and become regional nodes. In this case, the subset is no true subset and is just a "copy" of the original set of nodes.

[0014] Even if only the closest node is considered, a path with a complicated - in particular non-straight - structure, could lead to unwanted results like a "zigzagging" node sequence. In order to avoid such effects, it is preferred that if a waypoint is within the proximity zone of a node, this node is deleted from the subset of regional nodes. It will thus not be considered again for the node sequence of this path. This, of course, only refers to a deletion from the subset. The node, if fulfilling the relevant criteria, is added to the node sequence and will not be deleted from it.

[0015] There are numerous ways of determining which nodes are regional nodes. Of course, the area considered should be continuous and include start point and end point of the transport path. A simple and efficient way of filtering for relevant nodes is that a node is selected to be a regional node if it is within an ellipse having a start point and an end point of the path as focal points. This, of course includes the case of the ellipse being almost circular. The size of the ellipse - i.e. the sum of distances to the focal points - may be chosen in various ways, but it is preferred that the ellipse is not too "slim", i.e. the maximum width should be at least e.g. 20 % of the distance between start point and end point. Otherwise, too many nodes could be excluded beforehand, leading to unsatisfying results.

[0016] Another filter option is to include all nodes within a circle. The centre of the circle should be midway between the start point and end point and its diameter should be larger than the distance between the start and end point.

[0017] Normally, the path data defining the paths are generated by a software according to navigation algorithms known in the art, i.e. a given start point and an end point are specified by a user and the software determines an appropriate route. Thus, the path of the transport means is determined according to start and end point before the transport is carried out. The software may be executed on the same processing system as the compression method itself or it may be executed on an external system, which then provides the path to the processing system. Optionally, position sensors may provide position data referring to transport means and the paths may correspond to the position data. This is normally just for verification of the route actually taken by the transport means, e.g. in order to record a deviation from the originally planned route. In such a case, the position data are based on a position measurement by the sensors. Such position sensors usually are GPS sensors, which are on board a transport means. The respective position data can be stored on board the transport means and be transmitted after completion of the shipment. Alternatively, the position data can be sent to the processing system in real time using a transmitter on board the transport means. Such a transmitter could be provided by a smart phone having GPS capability. If an appropriate application for the smart phone is provided, it can provide position data and also send them e.g. using a mobile web technology. However, the transport means could also have "dedicated" components (sensors and transmitter) exclusively used for providing and transmitting position data to the processing system. Needless to say, another option is that the start and end point are specified by a user (e.g. a driver) in the transport means, and are transmitted to the processing system which determines an according route.

[0018] A main interest linked to the inventive method is to analyse the importance of a particular node combination, i.e. how much transport occurs between a given pair of nodes (or, strictly speaking, between the proximity zones of these nodes). It is preferred that for a pair of nodes, information regarding the paths of all node sequences which contain said pair of nodes is added up. Thus, for a given pair of nodes, e.g. "A" and "B", the method looks for all node sequences which contain "B" following "A" (or, alternatively, "A" following "B", because it should be distinguished which node occurs first) and adds up or accumulates information on the respective paths. In one embodiment, only those paths are taken into account where the pick-up point belongs to node A and the delivery point belongs to node B. One important option is that the times said pair of nodes occurs are counted. In this case, the information added up is just the integer "1 " for each relevant node sequence. The processing system, which can be referred to as an automatic learning system, provides a counter for each pair of nodes and every time this pair of nodes occurs in a node sequence, the counter is increased by 1 . The final value of the counter is a representation of the flow between the respective pair of nodes.

[0019] Alternatively, or additionally, a characteristic parameter of the transport can be provided and this characteristic parameter is added up. The characteristic parameter may be the volume, the weight or the value of goods transported. These parameters allow for an assessment how "important" a transport is. For some considerations, it is important to know not only how many transports go from "A" to "B" but also if small or large amounts of goods are transported.

[0020] A further possibility is to analyse the number of transport means needed for a specific node. In such a case, it may be counted how many times this node (i.e. the node plus its proximity zone) was used as a pick-up point (start point) for a transport. Such a count allows understanding fluctuation and regularity between various periods. This count may be done e.g. for every week of a year and this allows to forecast the need for transport means for a specific period. Such forecast values may again be compared with actual values to gain insight on the predictability. By carrying out trend comparisons between different periods (e.g. two subsequent weeks or years) the present method allows to predict the transport needs for a specific period in the future. [0021] A considerable potential of the method lies in discovering and utilising possibilities to perform reloading and/or co-loading. Such possibilities can arise when a path is near a location which is intended for a loading operation. Here, the term "loading" also includes unloading and reloading. There are many examples for such a loading operation. E.g. when a first transport passes near by a location which is intended as a pick up point for a second transport, the loading operation is the pick-up operation. In such a case, it may be considered to reroute the first transport to pick up the cargo, which may mean that both transports are performed by the same transport means. There are other examples for loading operations, as will be explained below. For such cases, it is preferred that, if a node sequence of a path comprises at least one node intended for a loading operation, an alternative path is planned which includes this node. I.e. the processing system plans an alternative path which enables a loading operation at this node. This may refer to a loading, unloading or reloading operation. Anyhow, the alternative path may be discarded by the processing system itself or by a user if it is considered unfavourable by some criteria.

[0022] The inventive method can be very useful in the optimum employment of cross- docks. It is often not economical to do a transport by a single run with one transport means from start point to end point. Instead, it can be better to reload goods at a cross- dock so that the transport is shared among several transport means, each of which only travels a moderate distance and is available again afterwards. In many cases, co-loading for different transports is possible. Therefore, it is preferred that if a node sequence comprises at least one node corresponding to a cross-dock, an alternative path is planned which includes said at least one node and corresponds to a cross-docking operation at said at least one node. This means that the method searches for possibilities to employ cross-docks, thus dividing a transport into parts that can be done by different transport means. In particular, several nodes of a node sequence may correspond to several cross- docks. In such a case, there are different possibilities to divide the transport, from using only one cross-dock to using all cross-docks. The processing system performing the method may discard some of these options because the distance between two cross- docks is too small or too great, both of which is uneconomical and may put at risk the quality of service. Constraints such as maximum driving hours are also taken into account. Also, combinations with too many cross-docks may be discarded, because the total cross- docking time (i.e. the total time for reloading) may become too long. Furthermore, cross- docking may need to occur within a given time window to enable co-loading. In other words, one reloading operation (e.g. truck A to truck C at cross-dock X) may have to be coordinated with another reloading operation (truck B to truck C at cross-dock X). It should be noted that even if an alternative path is planned as described above, it may be discarded - with or without prompting a user - e.g. because it is uneconomical.

[0023] In order to identify the most important cross-docks, the method may calculate a "recurrence factor" for each cross-dock, i.e. a value that represents how many times a cross-dock was used within a given time period. Such a recurrence factor may be based on a count of all paths that include a cross-dock. In this case, normally only those paths should be considered that involve a cross-docking operation and not those which only pass through the proximity zone of a cross-dock. Cross-docks with a high recurrence factor may be selected with preference for an alternative path as described above.

[0024] The abovementioned employment of cross-docks normally refers to a reloading between transport means belonging to the same logistics provider. I.e. the routes and/or departure times of the transport means may be chosen according to the logistics provider's choice. Another possibility for co-loading is to use transport means belonging to third parties, i.e. other logistics providers etc. Of course, schedule and route of such third party transports are fixed. However, employing a third party transport for co-loading may save resources considerably. Therefore, in another embodiment of the invention, if a node sequence comprises at least one node corresponding to a start point or end point of a third party transport, an alternative path is planned which includes said third party transport. Again, such a path may be planned but be discarded anyway and the original path may be favoured. Normally, the path with the third party transport is selected if the total length of the path to the start point of the third party transport (pre-shipping) plus the path from the end point of the third party transport (post-shipping) is smaller than the original path without the third party transport. In a simplified version, only air-line distances may be considered in this comparison. Furthermore, the alternative path may be discarded if its timing is not compatible with the intended transport.

[0025] When two transports (partially) take place in the same area and go in a similar direction, it is worthwhile to consider if these two could be combined to one transport. I.e. the corresponding transport means of one transport could make a detour to carry out the other transport as well. In order to discover such options, according to one embodiment start points and end points of paths are provided as nodes and, if a node sequence of a first path comprises a node corresponding to a start point or end point of a second path, an alternative path is planned which includes the start points and end points of the first path and the second path. This means that the transport corresponding to the alternative path covers the pick-up and delivery points of the transports corresponding to the first and the second path. In other words, one transport means could carry out both transports. [0026] There are numerous possible modifications to the inventive method. For instance, the number of nodes may be reduced, say to only one or two per country. Of course, in such a case, the proximity zones of the nodes should be large enough. In this case, it is possible to study in a very generalised way the flows from country to country. One could also analyse what the total flow from a specific node (i.e. from this node to any other node) and the total flow into this node is. Each of these flows gives information about the importance of this node. A comparison of these flows shows whether there is a balance or not. One could also analyse the balance between the times a node is used as a pick-up point and the times it is used as a delivery point. If it is used more often as a pick-up point, more transport means are needed there. The same analysis can be done for several nodes located in a single region, e.g. a country, to get a "regional balance". It is also possible to do such analysis separately for each of several logistics providers. Apart from showing the results as numbers, they could also be visualised in a map, where e.g. arrows of different sizes show flows, whereas "unbalanced" nodes or regions are highlighted by different colours etc. Analysis of the flows also makes it possible to find optimal positions for cross-docks. As mentioned above, the nodes can have individual proximity zones, e.g. having different radiuses or completely different shape. These are just examples and are not limiting to the applicability of the invention.

[0027] The invention also provides a processing system for compressing transport data by a mapping procedure using node data. The processing system is configured to be provided with the transport data, which define a plurality of paths on an at least two- dimensional map, each path describing a transport. This includes the possibility that only a start point and an end point for a path are provided and the system itself determines the path. The system is further configured to be provided with the node data, which define a set of nodes on the map, each node having a proximity zone. The processing system is also configured to perform, for each path, the mapping procedure to obtain sequence data defining a node sequence consisting of nodes through whose proximity zones the path passes. The terms used here have already been explained above with reference to the inventive method and will not be explained again. The processing system may also be referred to as a data compression system.

[0028] To be provided with the abovementioned data, the processing system usually needs some kind of interface. Further, for storing the received data, processing system comprises or at least has access to a memory device.

[0029] Preferred embodiments of the inventive processing system correspond to those of the inventive method. Brief Description of the Drawings

[0030] Preferred embodiments of the invention will now be described, by way of example, with reference to the accompanying drawings, in which:

Fig. 1 is a schematic view of a transport means and a processing system according to the invention;

Fig. 2 is a diagram illustrating a mapping procedure by the inventive method in a first stage;

Fig. 3 is a diagram illustrating the mapping procedure in a second stage;

Fig. 4a is a diagram illustrating a transport path and several cross-docks;

Fig. 4b is a diagram illustrating an alternative path for the transport path of fig. 4a;

Fig. 5a is a diagram illustrating a transport path and a third party transport;

Fig. 5b is a diagram illustrating an alternative path for the transport path of fig. 5a;

Fig. 6a is a diagram illustrating two transport paths; and

Fig. 6b is a diagram illustrating an alternative path for the transport paths of fig. 6a

Description of Preferred Embodiments

[0031] Fig.1 shows a processing system 40 according to the invention. It comprises a processing unit 41 (e.g. a CPU of a personal computer), which is connected to a memory device 42 (e.g. a main memory and/or hard disc of the personal computer). The processing unit 41 can read data from the memory device 42 and store data onto it. Further, the processing unit 41 is connected to a first interface 43 and a second interface 44 for receiving external data. The first interface 43 may be an input device like a keyboard or mouse or may be a connection to an external data source like a network. The second interface 44 may be configured to receive wired or wireless signals.

[0032] The processing system 40 is configured to receive path data referring to a plurality of paths 1 on a map. Each path 1 describes a transport carried out by a transport means 100, which is schematically shown in Fig. 1 as a truck. The path data may be input by means of the first interface 43. In some cases, the details of the path 1 may be generated by the processing system 40 itself. E.g. a start point 2 and an end point 8 are specified by a user input via the first interface 43 and the details of the path are determined by the processing system 40 by navigation algorithms known in the art. Of course, such navigation algorithms could also be executed externally.

[0033] The truck 100 is equipped with a GPS device 101 which determines its position. The position data provided by the GPS device 101 are sent by a wireless transmitter 102 to the second interface 44. Usually, the signal from the transmitter 102 is relayed using a terrestrial or satellite connection. Here, the GPS device 101 and the transmitter 102 are shown as dedicated components of the truck 100. However, they could also be part of a smart phone having GPS capability. The communication via the transmitter 102 and second interface 44 is mainly for verifying the position of the truck 100 and possibly for rerouting.

[0034] The data provided by the transmitter 102 corresponds to a sequence of coordinate pairs representing the path 1 of the transport. Since the smaller details of the path 1 are not important for the subsequent analysis, the (intended) position of the truck 100 will only be stored for every 10 km. Either the position data are only provided to the system 40 for every 10 km or the processing unit 41 determines these 10-km-steps and derives the corresponding waypoints - if necessary, by interpolation - from the original position data.

[0035] As a result, the path 1 is represented by path data corresponding to a series of waypoints 2-8 on a two-dimensional map as shown in Fig.2. The path 1 runs from a start point 2 (over intermediate way points 3-7) to an end point 8, which correspond to the points where the truck 100 picks up and delivers goods, respectively. On this map, there is further defined a set of nodes 10-22 by corresponding node data, each of which nodes 10-22 corresponds to a significant point, like a city, a cross-dock, a warehouse, a railway station, an airport etc. For each node 10-22, there is defined a circular proximity zone 30. In this case, all proximity zones have a radius equal to 15 km.

[0036] At the beginning of the analysis, the processing system 40 determines some regional nodes 10-19, which are considered to be in a relevant region of the path 1 . Only those nodes 10-19 are selected which are within an ellipse 31 having the start point 2 and the end point 8 as focal points. Nodes 20-22 outside the ellipse 31 may not be considered for this path.

[0037] The processing system 40 checks waypoint-by-waypoint if the path is in the proximity zone 30 of a regional node 10-19. The start point 2 and the first intermediate waypoint 3 are not within such a proximity zone 30. The next waypoint 4, however, is within the proximity zone of node 15. It is not within the proximity zone 30 of any other node 10-19, therefore node 15 is selected as the start of a node sequence. The selection is indicated in Fig. 2 by the black circle around node 15. Further, node 15 is deleted from the set of regional nodes 10-19, whence it will not be considered again for this path.

[0038] After two more steps, at another waypoint 5, the path enters the proximity zone 30 of another node 1 1. It is still within the proximity zone 30 of node 15, but this is no longer considered due to its deletion from the set of regional nodes. Therefore, node 1 1 is added to the node sequence and deleted from the regional nodes. After four more steps, at waypoint 6, the path 1 enters the proximity zone 30 of another node 17, which is also selected for the node sequence. After another four steps, at waypoint 7, the path 1 simultaneously enters the proximity zones of two nodes 13, 19. In this case, the processing system 40 selects the node 19, which is closest to the current waypoint 7. However, both nodes 13, 19 are deleted from the set of regional nodes.

[0039] No additional proximity zones are encountered on the way to the end point 8, whence the node sequence is completed. It consists of node 15, node 1 1 , node 17 and node 19. The mapping procedure for mapping the waypoint sequence of the path 1 to a node sequence has been completed.

[0040] The processing system 40 will store sequence data defining the node sequence on memory device 42. The same mapping procedure is done for a plurality of paths 1 , each of which corresponds to a route of a transport means 100. Afterwards, the processing system 40 generates a "flow matrix" in which each line and column corresponds to a node. Initially, all values of the matrix are zero. The processing system 40 checks all stored node sequences. If a node sequence contains e.g. consecutive nodes 1 1 and 17, the corresponding value in line 1 1 , column 17 of the matrix is increased by one. Of course, the matrix could also be generated as the node sequences are determined, even without explicitly storing the sequence data of any node sequence.

[0041] In any case, the resulting flow matrix allows for analysis of important features like the flow from one node to another, the overall flow from (to) one node to (from) all other nodes, the balance of flows to and from a node etc. It is also possible to include a characteristic value like the weight of the goods transported. In this case, a flow matrix would not include simple integer values but values which correspond to a total weight. If a node combination is included in a node sequence, the corresponding value in the matrix is increased by the weight of the goods of the respective transport.

[0042] Fig. 4a shows a transport path 1 , wherein, for sake of simplicity, only the start point 2 and end point 8 are shown while intermediate points are omitted. Also shown are three nodes 23-25 which represent cross-docks used by the logistics provider. As can be seen in the figure, the path 1 passes through the proximity zones 30 of two nodes 23, 24. These are included into the node sequence of the path 1 by the above-explained mapping procedure (possibly along with other nodes which are not shown here). The processing system 40 analyses the possibility of employing one or both of the cross-docks 23, 24 for reloading operations. For instance, an alternative path 1 b is planned as shown in Fig. 4b. This path 1 b is rerouted to include both cross-docks 23, 24. It is understood that such a path 1 b is normally carried out by three different transport means, one for each part. Also, the processing system will plan alternative paths which include only one of the two cross- docks 23, 24. The actual path used for the transport may be chosen e.g. depending on whether the transport means used between the two cross-docks 23, 24 can be used for at least one other transport, i.e. co-loading is possible. The choice may be made by the processing system 40 itself or by a user, optionally after pre-selection by the processing system 40.

[0043] Figs. 5a, 5b illustrate an example of employing third party transports by using the inventive method. A transport path 1 with its start point 2 and end point 8 is shown in Fig.5a, along with a node 26 representing the start point (i.e. pick up point) and a node 27 representing the end point (i.e. delivery point) of a third party transport. This third party transport, which is simply represented by a straight dashed line, is organised by e.g. another logistics provider. The route and schedule of this transport cannot be influenced by the logistics provider who organises the transport represented by path 1. Anyhow, it can be advantageous to employ the third party transport for a part of the transport from start point 2 to end point 8. The processing system 40 detects such a possibility by the path 1 entering the proximity zone 30 of node 26. Accordingly, an alternative path 1 b is planned as shown in Fig. 5b. This path 1 b is rerouted to include nodes 27, 28. It runs from start point 2 to node 26, where goods are reloaded to the third party transport, and from node 27 to end point 8. Whether the initial path 1 or the alternative path 1 b is used for the transport may depend e.g. sum of the distances (or associated costs) from start point 2 to node 26 and from node 27 to end point 8 is greater or smaller than the distance between start point 2 and end point 8. For simplification, air-line distances may be considered. Another criteria is of course the schedule of the third party transport.

[0044] Figs. 6a and 6b illustrate an example of co-loading without an intermediate change of transport means. Fig. 6a shows a first transport path 1 with its start point 2 and end point 8 and a second transport path 1 a with a start point 2a and an end point 8a. In this case, the start and end point 2a, 8a are treated as nodes in the mapping procedure of the first path 1. The same applies, vice versa, to the mapping procedure of the second path 1 a, which, however, is not illustrated here. The fact that the first path 1 passes through the proximity zone 30 of the start point 2a of the second path 1 a is taken as an indication that the transport represented by paths 1 , 1 a may potentially be combined. Accordingly, an alternative path 1 b is planned as shown in Fig. 6b. This path 1 b is rerouted to include both start points 2, 2a and both end points 8, 8a. It runs from start point 2 via start point 2a and end point 8a to end point 8. The processing system will also plan other paths (not shown) in which the sequence of the points 2, 2a, 8, 8a is different. In the present case, it is obvious that the overall distance of the alternative path 1 b is smaller than the sum of the paths 1 and 1 a. Other sequences may also lead to a reduction compared to the individual paths 1 , 1 a, but the sequence shown in Fig. 6a is the optimum.

[0045] It should be noted that embodiments of the inventive method shown in Figs. 4a-6b can be combined with each other, i.e. the inventive method may search for options to employ cross-docking, third party transports and to combine transports at the same time.

Legend of Reference Numbers:

1 , 1 a, 1 b path 41 processing unit

2, 2a start point 42 memory device

3-7 waypoint 43, 44 interface

8, 8a end point 100 truck

10-27 node 101 GPS device

30 proximity zone 102 transmitter

31 ellipse

40 processing system