Method of operating laboratory sample distribution system, laboratory sample distribution system and laboratory automation system

Abstract

The present disclosure relates to a method of operating a laboratory sample distribution system having: a plurality of carriers (4) configured to carry one or more sample containers containing a sample to be analyzed by a laboratory device (3); a transport plane (1) assigned to the laboratory device (3) and providing support to the plurality of carriers (4); and a drive device (13) configured to move the plurality of carriers (4) between planar orientations (5) provided on the transport plane (1) in response to a drive control signal. The method comprises: pre-determining, by one or more processors of a data processing device, an off-line route (6) on the transport plane (1) prior to moving the vehicle (4) on the transport plane (1), the pre-determining comprising: determining a model, the model represents the transport plane (1) having a position-to-position movement between a planar position (5 ') and a planar position (5') associated with the plurality of carriers (4); using said model to calculate an optimized set of off-line routes between pairs of planar positions from a plurality of planar positions (5 '), said calculation comprising solving an optimization problem in which routes between said pairs of planar positions are optimized simultaneously; and providing the optimized set of offline routes as an offline route (6) on the transport plane (1); and controlling the drive device (13) in order to move the carrier (4) along a predetermined off-line path (6) on the transport plane (1). In addition, the present disclosure provides a laboratory sample distribution system and a laboratory automation system.

Description

[0001] The present disclosure relates to a method for operating a laboratory sample distribution system. Further, the present disclosure relates to a laboratory sample distribution system. In addition, the laboratory sample distribution system is also referred to as a laboratory automation system. Background Art

[0002] Laboratory automation systems are particularly suitable for substantially automatically determining samples, such as body fluid samples.The samples are typically received in sample vessels or containers that are processed via the laboratory automation system.

[0003] Such laboratory automation systems can include multiple units. Laboratory automation systems typically include multiple laboratory stations or devices, such as, for example, pre-analysis, analysis and / or post-analysis stations or devices. Typically, containers are transported between different stations of the system via a sample distribution system or a transport system. Sample vessel carriers with or without samples can be moved along a line for processing samples, wherein the sample vessel carriers are moved by a transport device having one or more actuators and actuator drivers or drive devices for driving the carriers. For example, in a system, a sample vessel can be moved or repositioned from a first workstation to a second workstation disposed in a processing line. A workstation or device can also be referred to as a work position.

[0004] At present, only the transportation system in the laboratory is fully automatic. However, the design of these transportation systems is mostly quite simple, including a conveyor system, in which samples and sample containers / carriers move along fixed routes respectively. Usually, a set of routes along different analyzers / laboratory stations is defined via hardware design (placement of rails / tracks) and / or electronic equipment (DIP switch configuration, pre-programmed turntable logic, etc.) (for example, a belt-driven transportation system). Therefore, a specific sample / container / carrier can be assigned to a specific fixed route. A specific sample / container / carrier can be transported along this specific fixed route, can "step out" the route when approaching a specific analyzer to be processed by the analyzer, or can "step on" the route again to further transport along a specific fixed route until the end of the route is reached. Samples can then be stored or discarded, for example. Usually, these routes are designed by "brainpower" with a manual process, that is, for a specific order situation, such as a laboratory designer defines a route to efficiently process an order reference set and meet predefined requirements.

[0005] However, for complex conveying systems this design may reach its limits.

[0006] Document WO 2016 / 188 752A1 discloses a laboratory automation system, which includes: a certain number of laboratory stations, and a certain number of sample container carriers (e.g., 10 to 10,000), wherein the sample container carrier is adapted to carry one or more sample containers. The sample container includes a sample to be processed by means of the laboratory station. A transport plane is provided, wherein the transport plane is adapted to support the sample container carrier. The transport plane includes a certain number of transfer areas, wherein the transfer areas in the certain number of transfer areas are assigned to corresponding laboratory stations in a certain number of laboratory stations. By moving the corresponding sample container carrier to the assigned transfer area, the sample is transferred to the laboratory station. A drive member is provided, wherein the drive member is adapted to move the sample container carrier on the transport plane simultaneously and independently of each other along a separate transport path.

[0007] More complex delivery systems may require more sophisticated operating methods, especially in order to better realize the potential of the delivery system.

[0008] Document EP 3 095 739 A1 discloses a method for operating a laboratory sample distribution system, which includes a certain number of sample container carriers adapted to carry one or more sample containers, wherein the sample container includes samples to be analyzed by means of a certain number of laboratory stations. A transport plane is provided, wherein the transport plane is adapted to support the sample container carrier. The transport plane includes a certain number of transfer positions assigned to the corresponding laboratory stations. A drive member is provided, which is adapted to move the sample container carrier on the transport plane. The method comprises the following steps: during the initialization of the laboratory sample distribution system, a route is pre-calculated according to the transfer position; and after the initialization of the laboratory sample distribution system, the drive member is controlled so that the sample container carrier moves along the pre-calculated path. During the initialization (startup, booting) of the laboratory sample distribution system, a fixed route extending on the transport plane is pre-calculated according to different transfer positions (between different transfer positions). A pre-calculated route is provided between the transfer positions on the transport plane. The transfer position represents an initial or target node in the sense of graph theory. The route is calculated using an informed search algorithm (i.e., an A* algorithm or a D* algorithm). The A* algorithm is an algorithm for path finding and graph traversal, which is used to efficiently calculate traversable paths between different nodes (eg, in the form of transition locations).

[0009] Document EP 3 410 123 A1 relates to a method of operating a laboratory sample distribution system, wherein the laboratory sample distribution system comprises: a certain number of sample container carriers, wherein each of the sample container carriers comprises at least one magnetically active device and wherein each of the sample container carriers is adapted to carry at least one sample container; and a certain number of interconnected transport plane modules, wherein each of the number plane modules is adapted to support a certain number of said sample container carriers. A certain number of electromagnetic actuators are provided, wherein under each transport plane module, a certain number of said electromagnetic actuators are fixedly arranged in rows and columns. The electromagnetic actuators are adapted to move the sample container carriers of the sample container carriers along one of the rows or along one of the columns over the transport plane modules by applying a magnetic moving force to the sample container carriers. The method comprises the following steps: a) assigning at least one of the transport plane modules to a route category, wherein at least two flow channels are formed on the transport plane modules classified by routes, wherein the sample container carrier moves in each flow channel in a given transport direction, wherein the transport directions of at least the two flow channels are opposite to each other, and wherein for the sample container carrier moving on the transport plane module classified by routes, it is impossible to change from one transport direction to an opposite transport direction; and b) assigning at least one other of the transport plane modules to a waypoint category, wherein for the sample container carrier moving on the transport plane module classified by waypoints, it is possible to change from one transport direction to an opposite transport direction.

[0010] Document US 9 835 637 B2 generally relates to an automation system for a laboratory environment, and more specifically to a system and method for scheduling samples within an automation system by providing queuing logic.

[0011] An analytical system for analyzing biological samples is disclosed in document US2019 / 0 120 866A1.

[0012] Document US Pat. No. 10,668,622 B2 relates generally to an automation system for a laboratory environment and more specifically to a system and method for a clinical analyzer.

[0013] An automated system for an in vitro diagnostic environment is disclosed in document US 9 315 334 B2. The system includes a plurality of intelligent vehicles including onboard processing and navigation functions. The vehicles control local motion and navigate at decision points (such as forks in the track) to independently reach the appropriate test station.

[0014] With complex transport systems, the calculation of routes for vehicles also becomes more complex. The size of the set of possible routes grows exponentially with the number of vehicles and the number of transport zones of the transport plane / transport plane module. Even for a single vehicle, there are a large number of routes that connect the current position of the single vehicle to the target destination. However, in a typical use case (with a large number of transport zones and vehicles), the optimal routing should consider routes for hundreds of vehicles simultaneously. Summary of the invention

[0015] The object is to provide an improved method of operating a laboratory sample dispensing system and a laboratory sample dispensing system. In particular, the object is to provide a technique for improved routing of a carrier to a laboratory sample dispensing system.

[0016] To solve this problem, a method of operating a laboratory sample dispensing system is provided according to independent claim 1. Further, a laboratory sample dispensing system is provided according to independent claim 16. Additionally, a laboratory automation system is provided according to claim 17. Further embodiments are disclosed in the dependent claims.

[0017] According to one aspect, a method of operating a laboratory sample distribution system is provided. The laboratory sample distribution system includes: a plurality of carriers configured to carry one or more sample containers containing samples to be analyzed by a laboratory device; a transport plane assigned to the laboratory device and providing support to the plurality of carriers; and a drive device configured to move the plurality of carriers between planar orientations provided on the transport plane in response to a drive control signal. The method includes predetermining (based on the transfer position) an offline route on the transport plane by one or more processors of a data processing device before moving the carrier on the transport plane. Predetermining includes the following steps: determining a model representing a transport plane having planar positions and position-to-position movements between the planar positions associated with a plurality of carriers; using the model to calculate an optimized set of (multiple) offline routes between pairs of planar positions from the plurality of planar positions, the calculation including solving an optimization problem in which the routes between the pairs of planar positions are simultaneously optimized; and providing the optimized set of offline routes as offline routes on the transport plane (e.g., to the drive device). Controlling the drive device so as to move the carrier along the predetermined set of offline routes on the transport plane.

[0018] According to another aspect, a laboratory sample distribution system is provided. The laboratory sample distribution system includes: a plurality of carriers configured to carry one or more sample containers containing samples to be analyzed by a laboratory station; a transport plane assigned to a laboratory device and providing support to the plurality of carriers; and a drive device configured to move the plurality of carriers between planar orientations provided on the transport plane in response to a drive control signal. The laboratory sample distribution system is configured to predetermine an offline route on the transport plane by one or more processors of a data processing device (based on a transfer position) before moving the carrier on the transport plane. Predetermining includes: determining a model representing a transport plane having planar positions and position-to-position movements between the planar positions associated with a plurality of carriers; using the model to calculate an optimized set (multiple) of offline routes between pairs of planar positions from the plurality of planar positions, the calculation including solving an optimization problem in which the routes between the pairs of planar positions are simultaneously optimized with respect to each other; and providing the optimized set of offline routes as offline routes on the transport plane. The laboratory sample distribution system is further configured to control the drive device to move the carrier along a predetermined optimized set of off-line routes on the transport plane.

[0019] According to a further aspect, a laboratory automation system is provided, comprising the laboratory sample dispensing system according to the preceding aspect and a plurality of laboratory devices.

[0020] The predetermined offline route may correspond to at least one of the following: a route that may be determined when designing the configuration (laboratory sample distribution system); a route that is determined when the system is turned on; a route that is manually determined, for example, triggered by an operator; and a route that is triggered by the system itself (for example, when the flow / order / instrument or the state of a portion of the transport surface changes). In particular, the predetermined route may correspond to a route that is not calculated for each individual carrier but is fixed within a certain time (period).

[0021] As used, the term "offline" refers to a route that is calculated offline. An offline route may be a route that is not calculated in real time for each movement when operating the laboratory sample distribution system, but rather, for example, during design of the laboratory or at the beginning of a workday before actually running or operating the laboratory sample distribution system (runtime).

[0022] Calculating an optimized set of offline routes between pairs of plane locations from multiple plane locations may further include: determining multiple offline routes between pairs of plane locations from multiple plane locations (e.g., in a directed graph model), and determining an optimized set of offline routes from the multiple routes, including solving an optimization problem (e.g., a mixed integer optimization problem) for the multiple offline routes.

[0023] The model can be a directed graph model of a transport plane, wherein the nodes of the directed graph model are assigned with plane positions, and the arcs connecting the nodes of the directed graph model are assigned with (feasible) position-to-position movements between two plane positions. Alternatively, the model can be a model from graph theory, a numerical (statistical) model and / or a model used in simulations. However, the model can also be another model, such as a combination of the above models.

[0024] The plane position may include one or more (a small group of) plane orientations. Each plane orientation (of the transport plane) may correspond to a certain area / logical area on the transport plane. The transport plane may be divided into several logical areas, such as square logical areas of the same size and outline. The logical areas may form a regular grid, wherein each cell of the grid may correspond to a logical area, each cell of the grid may be a square of the same size (the same in the X and Y directions), and / or the grid may (precisely) cover the transport plane. One / each transport module (TM) may include one or more logical areas (e.g., a subset of logical areas forming a transport plane). A group of transport modules may form a transport plane. The pitch between adjacent transport modules (TM) may form a pitch in a regular grid. For example, 10 to 10,000 plane orientations may be provided. Each plane orientation / logical area may correspond to a transport module, and a plurality of transport modules together constitute a transport plane. The nodes of the directed graph model may be assigned one or more (a small group of) logical areas / plane orientations. Arcs connecting nodes of the directed graph model may be assigned location-to-location movement between two logical region / plane location groups, where each group may include one or more logical regions / plane locations.

[0025] The optimized set of off-line routes provides a predetermined off-line route on the transport plane. In an embodiment, location-to-location movement may be allowed only between two adjacent plane locations. Location-to-location movement may be allowed only between two adjacent groups of logical zones / plane locations. Arcs may only connect adjacent nodes of a directed graph model.

[0026] Multiple off-line routes between plane position pairs from multiple plane positions in the directed graph model can be selected to form multiple possible routes for connecting the plane position pairs. Possible routes can be (directly) located on the transport plane, preferably completely (directly) located on the transport plane. Each plane position pair can indicate a pair of starting position / node and destination position / node of a route in the route. The starting and destination positions of a certain route can constitute the endpoint positions of the route. The plane position pair can correspond to the transfer position pair.

[0027] The planar position pairs may correspond to the most frequent starting and destination positions (endpoint positions) of the vehicle.

[0028] The transfer position pairs and / or plane position pairs can represent the corresponding pairs of the starting node and the destination node of the route in the graph theory sense. The route can be predetermined between the transfer positions, i.e. the end point or end node of the route is formed by the transfer positions. If, for example, the first, second and third (most frequent) end point positions of the vehicle are given, two routes (in one direction and in another direction) between the first and second (most frequent) end point positions, two routes (in one direction and in another direction) between the first and third (most frequent) end point positions and two routes (in one direction and in another direction) between the second and third (most frequent) end point positions can be predetermined.

[0029] For example, 10 to 10,000 sample container carriers may be provided. The laboratory sample distribution system may process 2,000 to 200,000 samples per day. For example, 2 to 50 laboratory stations may be provided. The transport plane may be completely or at least partially planar.

[0030] A first and a second (sub) transport plane may be provided, wherein the first and the second transport plane may be provided at different levels. The carrier may be transported from the first transport plane to the second transport plane (or vice versa) via an elevator (e.g., a paternoster) and / or a ramp. The transport plane may include the first and the second (sub) transport plane and optionally further (sub) transport planes. Therefore, the method for determining an offline route may also include a route extending over both the first and the second transport plane and optionally further transport planes, each of which is provided at a different level.

[0031] In this regard, a route connecting segment of a route extending across multiple transport planes (each provided at a different level) may connect a route segment on one of these planes (e.g., a first transport plane) with a route segment on another of these planes (e.g., a second transport plane), wherein the route segment connecting the route segments at different levels will become part of the route. The route connecting segment may be included by a transport system (such as a ramp or an elevator).

[0032] In the present application, the term "conveying plane" generally refers to a two-dimensional and / or three-dimensional surface, i.e. both a planar surface and / or a surface extending not only in two spatial dimensions but also in a third spatial dimension. In the sense of the present application, such a three-dimensional surface can be, for example, a surface with one or more slopes between (sub-)planes, a ramp and / or an elevator.

[0033] The (sample container) carrier can be placed on the top of the transport plane. The carrier can move on and above the top of the transport plane. The transport plane can include 2 to 100 transfer positions. The transfer position can be assigned to the corresponding laboratory station or device. In addition or alternatively, the transfer position can be assigned to the area managed by the separate routing system. For example, the queue area can be controlled by the separate routing system. In addition, there can be two or more separately controlled transport systems that require switching (i.e., transfer mechanism) between these areas. For example, each laboratory station or device can have a single corresponding transfer position, which is configured to transfer the sample carrier and / or sample from the transport plane to the laboratory device or return. Alternatively, more than one transfer position can be assigned to the corresponding laboratory station or device, particularly the output transfer position and the input transfer position. The transfer position can be assigned to the laboratory station statically or dynamically. In other words, during operation / at runtime, if necessary, the transfer position can be changed.

[0034] The optimization problem that may be associated with the model may be one of the following: a multi-object network flow problem, particularly a multi-object network flow problem in a directed graph; a shortest path problem; and a minimum flow problem.

[0035] Computing an optimized set of offline routes between pairs of plane locations from multiple plane locations may include, for example, finding an optimal multi-item network flow in a directed graph. The multi-item network flow problem is a network flow problem in terms of multiple items (flow requirements) between different source nodes and sink nodes. The endpoint location / location pair may define a source node and a sink node.

[0036] The optimization problem can be solved, for example, by applying a MIP solver (mixed integer programming solver). A general solver for MIP problems can be used. The MIP solver can be one of the following: Gurobi, IBM ILOG Cplex, and Coin-OR CBC, i.e., the solution to the mixed integer optimization problem can be found using software products such as Gurobi, IBM ILOG Cplex, and Coin-OR CBC. There are known additional solvers such as that are capable of solving such optimization problems. In general, mixed integer optimization problems deal with mathematical optimization problems with two types of variables: variables taking values ​​in the integer domain, and variables taking values ​​in the continuous domain.

[0037] First, an optimization problem, such as a multi-object network flow problem, can be defined, that is, a purpose problem is modeled. The purpose problem can be to find a set of routes that are optimal with respect to one or more given criteria (constraints) on a transport plane. Then, a solution method for the defined optimization problem, such as a MIP solver, can be selected to solve the optimization problem.

[0038] Examples of possible solutions are presented in the P.FESTA paper "A brief introduction to exact, approximation, and heuristical algorithms for solving hard combinatorial optimization problems", 16th International Conference on Transparent Optical Networks (ICTON), 2014, pp. 1-20, doi: 10.1109 / ICTON.2014.6876285. The optimization problem can be solved by an exact optimization method, by approximating the original problem to a simpler problem and solving the simpler problem, and / or by a heuristic method and / or a heuristic solution method. The exact optimization method can be (i) a branch and bound method, (ii) a dynamic programming method, or (iii) a solver (other exact optimization methods can also be used). The solver can include a variety of algorithms (not just exact algorithms). The exact optimization method can also include a combination of the above methods. Solving the approximated simpler problem can include applying (i) one or more greedy algorithms, (ii) a local search, (iii) one or more relaxation-based algorithms, or (iv) one or more randomized algorithms (other methods for solving approximated simpler problems can also be used). Solving the simpler problem of approximation may also include a combination of the above methods. The heuristic method and / or heuristic solution may be (i) simulated annealing, (ii) one or more evolutionary algorithms, (iii) tabu search, or (iv) one or more greedy randomized adaptive search processes (GRASP). The heuristic method and / or heuristic solution may also include a combination of the above methods.

[0039] The method may further include determining a re-optimized set of offline routes from the optimized set of offline routes at the time of initializing the laboratory sample distribution system. Determining the re-optimized set of offline routes may depend on the expected vehicle flow on the transport plane. For example, the optimized set of offline routes may include two sets of optimized routes, the first set for high flow and the second set for low flow. If a lower flow is expected, the second set of optimized routes is selected when determining the re-optimized set of offline routes, and vice versa. The optimized set of offline routes may alternatively include several sets of optimized routes for several traffic scenarios. When determining the re-optimized set of offline routes, in this case, a corresponding set of optimized routes is selected from the several sets of optimized routes according to the expected flow.

[0040] The (expected) vehicle flow (vehicle flow intensity) may depend on several factors. The throughput on weekdays may be higher than on weekends, and the throughput between 8:00 and 18:00 may be higher than at night. For example, in the case of glucose screening of a batch of samples or clinical research, the type of tests ordered may be different on certain days or hours. Therefore, the endpoint (starting point or destination point) assigned to the corresponding laboratory device can serve as an endpoint more frequently. The corresponding route including the endpoint can have a higher vehicle flow (vehicle flow intensity). Similarly, due to seasonal effects or pandemic situations, the ordering of some tests will be more frequent over a longer period of time.

[0041] The determined optimized set of offline routes can be used as a lookup table or represented by the lookup table. Each time the vehicle needs to travel from the starting area to the destination, the corresponding route can be selected from the lookup table.

[0042] The method may further include, in a data processing device: (i) providing first frequent endpoint position data indicating a first selection of plane positions of endpoints that most frequently provide a travel route for the vehicle; and (ii) determining a directed graph model of the transport plane, wherein a first node of the directed graph model is assigned a plane position from the first selection of plane positions, and a first arc starting and / or ending at the first node of the directed graph model is assigned a position-to-position movement from and / or to a plane position from the first selection of plane positions.

[0043] Alternatively or in addition, the method may include providing in the data processing device first frequent endpoint position data / data about first frequent endpoint position pairs, the data indicating a first selection of plane positions that most frequently provide endpoints / a first selection of plane position pairs that most frequently provide endpoint pairs (origin and destination) for travel routes for the vehicle. The first selection of plane position pairs may correspond to 5 to 10 endpoint pairs, i.e., the first selection of plane position pairs may correspond to 5 to 10 travel routes for the vehicle.

[0044] The plane position pairs from multiple plane positions may include a first selection of plane positions and / or a first selection of plane position pairs. In particular, the first selection of plane positions and / or the first selection of plane position pairs may define plane position pairs from multiple plane positions. The first selection of plane positions and / or the first selection of plane position pairs may correspond to transfer positions and / or transfer position pairs (but not limited to this). However, the first selection of plane position pairs may correspond to all endpoint pairs (starting point and destination) for the vehicle's route of travel.

[0045] Alternatively or in addition, the method may include providing data about a first endpoint position and / or a first endpoint position pair in a data processing device, the data indicating a first selection of plane positions and / or a first selection of plane position pairs located in a first area, corresponding to a first workflow, and / or corresponding to high priority samples (time critical tests).

[0046] A transfer position can correspond to one or more than one laboratory device, such as an analyzer. For one or more than one laboratory device, multiple plane positions (each) can define a transfer position. A first plane position in a plurality of plane positions can define a transfer position. Alternatively, a second plane position in a plurality of plane positions can define a transfer position. It can be exchanged whether the first plane position or the second plane position defines a transfer position. In particular, during the predetermination of an offline route, in particular during the calculation of an optimized set of offline routes, it can be exchanged whether the first plane position or the second plane position defines a transfer position. The laboratory device can include an input for a carrier (assigned to the first transfer position) and an output for the carrier (assigned to the second transfer position). In particular, during the predetermination of an offline route, in particular during the calculation of an optimized set of offline routes, inputs and outputs can be exchanged. In particular, the exchange can be performed in one or more of the optimization steps. If the exchange has a performance advantage, the exchange can be performed. Only input and output pairs can correspond to plane position pairs. However, the input and output can correspond to the same plane position.

[0047] The method may further include, in a data processing device: (i) providing second frequent endpoint position data indicating a second selection of plane positions that less frequently provide endpoints of a travel route for the vehicle, wherein the second selection of plane positions is different from the first selection of plane positions; and (ii) determining a directed graph model of the transport plane, wherein a second node of the directed graph model is assigned a plane position from the second selection of plane positions, and a second arc starting and / or ending (connecting the second node) with the second node of the directed graph model is assigned a position-to-position movement from and / or to a plane position from the second selection of plane positions.

[0048] Alternatively or in addition, the method may include providing second frequent endpoint position data / data about second frequent endpoint position pairs in the data processing device, the data indicating a second selection of plane positions that less frequently provide endpoints / a second selection of plane position pairs that less frequently provide endpoint pairs (origin and destination) for the vehicle's travel routes. The second selection of plane position pairs may correspond to 5 to 10 endpoint pairs, i.e. the second selection of plane position pairs may correspond to 5 to 10 travel routes for the vehicle. The second selection of plane positions may be different from the first selection of plane positions (completely / in pairs). The second selection of plane position pairs may be different from the first selection of plane positions (completely / in pairs).

[0049] The plane position pairs from a plurality of plane positions may include a second selection of plane positions and / or a second selection of plane position pairs. In particular, the second and first selections of plane positions together and / or the second and first selections of plane position pairs together may define plane position pairs from a plurality of plane positions. The second selection of plane positions and / or the second selection of plane position pairs may correspond to non-transferred positions, i.e., the second selection of plane positions and / or the second selection of plane position pairs may not include transferred positions (but not limited to this).

[0050] Alternatively or in addition, the method may further comprise providing in the data processing device data regarding second endpoint positions and / or second endpoint position pairs, the data indicating a second selection of plane positions and / or a second selection of plane position pairs located in a second area, corresponding to a second workflow, and / or to samples having a lower priority.

[0051] The first selection of plane positions / plane position pairs may define a first set of plane position pairs. The second (and first) selection of plane positions / plane position pairs may define a second set of plane position pairs. The first and second sets of plane position pairs may be included by plane position pairs (may define the plane position pairs). The predetermination of the offline route may include: a first predetermination of the offline route, wherein the first set of plane position pairs defines the plane position pairs for the first predetermination; and a second predetermination of the offline route, wherein the second set of plane position pairs defines the plane position pairs for the second predetermination. The first and second predeterminations of the offline route may be performed independently of each other. The first and second (optimized) offline routes may correspond to the first and second predeterminations of the offline route. The first and second predeterminations of the offline route may correspond to different or the same optimization problems. The first and second predeterminations of the offline route may correspond to different or the same methods for solving the respective optimization problems. The predetermination of the offline route may include a third predetermination of the offline route. The third predetermination of the offline route may correspond to a third (optimized) offline route. The third predetermination of the offline route may include calculating an optimized set of offline routes between the third selection of plane positions / plane position pairs. Similarly, the fourth, fifth and further optimized offline routes may also be calculated in small groups (5 to 10 routes and / or 5 to 10 plane position pairs). The second predetermination of the offline routes may be performed after the first predetermination of the offline routes. The second predetermination of the offline routes may depend on the first (optimized) offline routes determined via the first predetermination of the offline routes. The expected traffic load from the first group of routes may be taken into account during the second optimization. During the second predetermination of the offline routes, the first (optimized) offline routes may be fixed. The first and second predetermination of the offline routes may correspond to determining a plurality of offline routes between plane position pairs from a plurality of plane positions in the directed graph model.

[0052] Optimizing offline routes in a group of several routes can make it easier to find a solution. For example, the optimization problem can be simplified in this case. In addition, the time required to perform the optimization procedure can be reduced. The optimization process can first focus on the most intensively used routes, while having lower optimization requirements for less intensively used routes.

[0053] It is also foreseeable to provide several routes (alternative routes) for the same origin-destination combination. This provides the option of switching between alternative routes in the event that one of the alternative routes has a problem, such as a traffic jam (the vehicle may be stuck on this alternative route). In addition, by using multiple routes (ideally using different logical positions), the traffic load on each route can be reduced to increase the total throughput (corresponding to a greater total capacity between origin-destination combinations) or to reduce wear on the surface used for the routes, because the traffic is distributed over more transport planes.

[0054] A pair in the first set of plane position pairs may include two plane positions from a first selection of plane positions. A pair in the second set of plane position pairs may include two plane positions from a second selection of plane positions. However, additionally or alternatively, a pair in the second set of plane position pairs may include one plane position from the first selection of plane positions and one plane position from the second selection of plane positions.

[0055] The method may further comprise, in a data processing device: (i) providing flow data indicating a predicted number of vehicles travelling between a pair of plane locations within a time interval; and (ii) calculating an optimized set of offline routes between the plane location pairs from a plurality of plane locations based on the predicted number of vehicles travelling between the plane location pairs. Additionally or alternatively, the flow data may indicate an actual / recent number of vehicles travelling between the plane location pairs within a time interval.

[0056] Providing flow data may include at least one of: providing flow data determined from a sample order list; providing flow data determined from historical data indicating historical operation of a laboratory sample dispensing system; providing flow data determined from workflow data indicating a workflow for one or more sample containers to be carried by a carrier; providing flow data determined from a measured current and / or recent number of carriers shipped; and providing flow data determined from a simulation. The flow data may correspond to empty and filled (routed) carriers.

[0057] Control of the drive device may further include: (i) in the drive device, receiving a reservation request from a vehicle that is traveling on a selected offline route from a predetermined offline route and is located at a current route position along the selected offline route, the reservation request indicating a request to reserve a subsequent route position (e.g., a logical area / plane orientation) along the selected offline route; (ii) verifying whether the subsequent route position is free for the drive device to travel; and (iii) if the subsequent route position is verified by the drive device to be free for travel, moving the vehicle from the current route position to a subsequent route position along the selected offline route.

[0058] After verifying that the subsequent route position is free for travel and before the vehicle is moved, the subsequent route position can be reserved for the vehicle, that is, the subsequent route position can be blocked for other vehicles. In particular, the subsequent route position can be blocked for other vehicles until the vehicle in question has arrived at and then left the subsequent route position. The above can be similarly applicable to n subsequent route positions. The n subsequent route orientations can correspond to a subgroup of orientations (e.g., logical area / plane orientations) along a selected offline route. In this case, the n subsequent route positions can be unlocked for the vehicle one by one after other vehicles have passed through the subsequent route position. Alternatively, the n subsequent route positions can also be unlocked for the vehicle immediately after other vehicles have passed through all of the subsequent route positions.

[0059] The calculation of an optimized set of offline routes between pairs of plane positions from a plurality of plane positions via solving an optimization problem may further include applying at least one constraint selected from the group consisting of: (i) minimizing the route length of each of the offline routes from the plurality of offline routes; (ii) minimizing the weighted route length of each of the offline routes; (iii) minimizing the number of route curves for each of the offline routes from the plurality of offline routes; (iv) minimizing the number of offline routes that join and / or intersect with another offline route; (v) evenly distributing vehicle traffic for each plane position; (vi) limiting position-to-position movement between two plane positions to movement only between adjacent plane positions; (vii) excluding plane positions reserved for vehicle queuing; (viii) evenly distributing predicted wear of plane positions on the plane positions of the transport plane; (ix) minimizing energy consumption of the laboratory sample distribution system; and (x) minimizing / avoiding an area of ​​2x2 plane orientations with four intersections. Minimize or avoid an area of ​​2x2 plane orientations that allow a cyclic sequence of feasible zone-to-zone movement. A region of a 2x2 plane orientation with four intersections may indicate a potential deadlock situation. Similarly, additional potential deadlock situations may be defined. Additional or alternative constraints may be to minimize or avoid one or more additional potential deadlock situations.

[0060] With respect to minimizing weighted route length, in an optimization process, weights may be assigned to pairs of endpoints, or an optimal route may be determined in a weighted graph.

[0061] In addition, constraints may have weights. Routes with higher expected traffic may have higher weights. Routes with higher weights may be preferred over routes with lower weights in terms of minimizing the corresponding route length, the number of route curves, and / or the number of intersections and / or overlaps with other routes (etc., see constraints above). The first and second predeterminations of the offline route may include applying different constraints.

[0062] In a typical scenario, for example, 5 to 10 routes between plane position pairs can be mutually optimized simultaneously. Each route can refer to a plane position pair. There can be 5 to 10 pairs of plane position pairs. Routes between 5 to 10 plane position pairs can be mutually optimized simultaneously.

[0063] Constraints can have weights. Weights can be set empirically. Constraints can be hard (e.g., can have "infinite weight") or soft (e.g., can have finite weight). Constraints associated with routes with high traffic can have a higher rating than routes with low traffic. According to one example (which refers to routes with lower traffic intensity), the following weights are applied in the optimization: (i) Route length: 5 penalty points for each zone (plane position or plane orientation) used in the route; (ii) Route joining: 50 penalty points for a route that joins with another route. According to another example (which refers to routes with higher traffic intensity), the following weights are applied in the optimization: (i) Route length: 10 penalty points for each zone (plane position or plane orientation) used in the route; (ii) Route joining: 100 penalty points for a route that joins with another route.

[0064] The predetermination of the offline routes may further include: in the data processing device, receiving first route traffic information indicating high vehicle traffic for a first offline route, and splitting at least one offline route into two or more different offline routes.

[0065] Determining multiple offline routes between plane location pairs from multiple plane locations in a directed graph model may further include: in a data processing device, receiving first route traffic information indicating high vehicle traffic for a first offline route, and splitting at least one offline route into two or more different offline routes.

[0066] Calculating an optimized set of offline routes between pairs of plane locations from the plurality of plane locations may further include receiving first route traffic information indicating high vehicle traffic for a first offline route, and splitting at least one offline route into two or more different offline routes.

[0067] The first route flow information can be derived by flow (intensity) data, which indicates the predicted number of vehicles traveling between the plane position pairs in a time interval. A (flow) threshold can be predefined. If the number of vehicles traveling between the (corresponding) plane position pairs in a time interval is greater than the threshold, the flow intensity can be identified as high (high flow intensity). If the number of vehicles traveling between the (corresponding) plane position pairs in a time interval is lower than the threshold, the flow intensity can be identified as low (low flow intensity).

[0068] Several (flow) thresholds of different sizes may be defined. The more (flow) thresholds are exceeded, the more frequently at least one offline route may be split. The at least one offline route may be split into as many different offline routes as the thresholds are exceeded. The (flow) threshold may indicate a maximum capacity of a route. In case there are several thresholds, a first / smallest threshold may indicate a maximum capacity of a route, whereas subsequent thresholds may indicate twice the maximum capacity of a route.

[0069] Different offline routes may include first and second offline routes that are different from each other. The first and second offline routes may have overlapping sections and spaced-apart sections. For example, the spaced-apart sections may include parallel sections. A vehicle that initially follows the route before splitting at least one offline route may follow one of two or more different offline routes into which the at least one offline route is split. The vehicle may be equally divided between two or more different offline routes. Therefore, in the case of two different offline routes, 50% of the vehicles may follow the first of the two different offline routes, and 50% of the vehicles may follow the second of the two different offline routes. Alternatively, the loads of different offline routes may be different. For example, the shortest of the different offline routes may have the highest load. If one of the different offline routes is optimal in terms of optimization (e.g., is the shortest and / or has the fewest intersections with other routes), the route may have the highest load among the different offline routes. Alternatively, one of the different offline routes may be loaded with traffic until a (traffic) threshold is reached, and all traffic exceeding the threshold may be sent to another route among the different offline routes.

[0070] The traffic intensity matrix or two-dimensional list may indicate the (expected) traffic intensity for each plane position pair (starting point / end point combination) (see Table 1). Each plane position pair may correspond to a certain (required) traffic intensity.

[0071]

[0072] Table 1: An example of a traffic intensity matrix, where origins are in columns and destinations are in rows. Each table entry represents the expected traffic intensity for the corresponding origin and destination pair.

[0073] The method may further comprise providing in the data processing means traffic intensity data indicating a predicted number of vehicles travelling between each of the pairs of planar positions within a time interval. The traffic intensity data may be comprised by the traffic data. The traffic intensity may comprise a traffic intensity matrix or a two-dimensional list.

[0074] Traffic data can be determined based on traffic intensity data and / or traffic intensity of each plane position pair. Traffic intensity data and / or traffic intensity of each plane position pair can be calculated according to the total cumulative traffic between each plane position pair, the peak traffic between each plane position pair and / or the capacity of the laboratory device assigned to each plane position pair. In order to determine the total cumulative traffic, all movements between each plane position pair can be calculated, such as from an order list, such as within a specific time interval. For example, the number of vehicles that move / estimate to move within 24 hours between each plane position pair can indicate the total cumulative traffic between each plane position pair. In order to determine the peak traffic, a time window (e.g., a moving time window) or a fixed time interval (e.g., 15 or 30 minutes within 24 hours) can be defined, wherein the maximum traffic intensity can be identified for each plane position pair. Peak traffic can correspond to global or local peak traffic. Global peak traffic corresponds to the traffic intensity of each plane position pair when the traffic between all plane position pairs is the largest. Local peak traffic corresponds to the maximum traffic intensity of each plane position pair. The maximum traffic intensity of different plane position pairs can occur at different times. The capacity of the laboratory device can be considered to avoid queues approaching the laboratory device. The sum of the flows (intensities) (for each route) may correspond to the transport capacity (for each route) of the system. The transport capacity (for each route) may be matched to the capacity of the laboratory device (for each route).

[0075] The above (flow) threshold value can be compared with the flow intensity data of each plane position pair to decide whether the flow is high.

[0076] The first selection of plane positions may indicate frequent plane position pairs (endpoints of vehicle routes). The second selection of plane positions may indicate less frequent plane position pairs (endpoints of vehicle routes). The first selection of plane positions may correspond to high entries in the traffic intensity matrix. The second selection of plane positions may correspond to low entries in the traffic intensity matrix.

[0077] The predetermination of the offline route may comprise: a first predetermination of the offline route, wherein the first selection of plane positions defines the plane position pairs; and a second predetermination of the offline route, wherein the second selection of plane positions defines the plane position pairs. The first predetermination of the offline route may be performed with priority. The first predetermination of the offline route may be performed before the second predetermination of the offline route. During the second predetermination of the offline route, the first predetermination of the offline route may be fixed / "frozen". The first and second predetermination of the offline route may be provided similarly to the predetermination of the offline route.

[0078] In the optimization, the flow intensity can be used to weight the importance of each pair of plane position pairs. For example, for plane position pairs with high flow intensity, it is more important to reduce the number of bends or intersections of the corresponding routes connecting the plane position pairs than for plane position pairs with lower flow intensity.

[0079] The predetermination of the offline route may further include: in the data processing device, receiving second route flow information indicating high vehicle flow (intensity) for the second offline route, and preventing the second offline route from being route adjusted when determining a plurality of offline routes and / or determining (calculating) an optimized set of offline routes. Alternatively or in addition, this step may be included in the step of calculating an optimized set of offline routes.

[0080] Calculating an optimized set of offline routes between plane position pairs from multiple plane positions using a model (e.g., in a directed graph model) may further include, in a data processing device: receiving first vehicle flow information indicating a first vehicle flow scenario for multiple offline routes; determining a first plurality of offline routes between plane position pairs from multiple plane positions (e.g., in the directed graph model); receiving second vehicle flow information indicating a second vehicle flow scenario for multiple offline routes, wherein the second vehicle flow scenario is different from the first vehicle flow scenario; and determining a second plurality of offline routes between plane position pairs from multiple plane positions (e.g., in the directed graph model). Alternatively or in addition, this step may be included by a predetermined step.

[0081] A first plurality of off-line routes between plane position pairs may include a first number of plane positions and / or plane orientations (logical zones) included / traversed by the first plurality of off-line routes. A second plurality of off-line routes between plane position pairs may include a second number of plane positions and / or plane orientations (logical zones) included / traversed by the second plurality of off-line routes. The first and second numbers of plane positions and / or plane orientations (logical zones) may be different. Each plane position and / or plane orientation (logical zone) may correspond to a corresponding module of the transport plane. Modules that only correspond to / include plane positions and / or plane orientations (logical zones) that are not included / traversed by the (first and / or second plurality) of off-line routes may be closed.

[0082] The control of the drive device may further include: operating the laboratory sample distribution system at runtime; and selecting an offline route from an optimized set of offline routes in the event that a runtime route cannot be determined for the vehicle at runtime. The control of the drive device may further include: operating the laboratory sample distribution system at runtime; and selecting an offline route from a lookup table. In this case, the offline route may serve as a backup route.

[0083] At runtime, the route for the vehicle can be (primarily) selected from the set of offline routes / lookup tables. However, one or more (single) routes can be re-optimized during runtime. Re-optimization of one or more (single) routes during runtime can be performed in response to user input, errors obtained for the route, or changes in corresponding endpoints. Re-optimization of one or more (single) routes during runtime can depend on the current route, such as a (predetermined) set of offline routes and / or (un)available plane positions (blocked plane positions). One or more (single) routes can be determined during runtime. At runtime, the routes determined offline and at runtime can be combined. Similarly, at runtime, a portion of the route can be re-optimized. Re-optimization can be performed in response to changes in the system (new laboratory devices, repositioning of laboratory devices, reconstruction of shipping planes, new workflows, new order profiles, defective modules of corresponding routes crossing shipping planes, and / or new vehicles).

[0084] Similar to the predetermination of the offline route, the offline route can be determined in the post-determination step during the movement of the carrier. The post-determination can be performed offline separately from the sample distribution system. The post-determination can depend on parameters provided by the sample distribution system and / or parameters provided by, for example, a user. The post-determination can be performed in response to changes in the system (new laboratory devices, repositioning of laboratory devices, reconstruction of the transport plane, new workflows, new order profiles, and / or new carriers). The offline route determined offline via the post-determination can then be loaded to the sample distribution system, that is, the offline route determined offline via the pre-determination can be replaced by the offline route determined offline via the post-determination.

[0085] In the method, predetermining the offline routes may further include: determining an optimized first set of offline routes; assigning a first application parameter to the optimized first set of offline routes; determining an optimized second set of offline routes that is different from the optimized first set of offline routes; and assigning a second application parameter to the optimized second set of offline routes. Further, controlling the drive device may further include: receiving application information indicating current application parameters; and selecting one of the optimized first set of offline routes and the optimized second set of offline routes to control the drive device if it is determined that the current application parameters match the first application parameters or the second application parameters. The first and second application parameters may indicate first and second dates, time periods, traffic conditions (e.g., high or low traffic), operation modes, and / or orders.

[0086] The application parameters may correspond to KPIs (Key Performance Indicators) of the laboratory automation system. Examples of application parameters may be flow data, constraints, weights of constraints, etc., or a combination thereof.

[0087] For a laboratory automation system, the plurality of laboratory devices (stations) may include one or more laboratory devices selected from: a laboratory device for pre-analysis; a laboratory device for sample analysis; and a laboratory device for post-analysis. The pre-analysis workstation may be adapted to perform any kind of pre-processing on a sample, a sample container, and / or a sample container carrier. The analysis workstation may be adapted to use a sample or a portion of a sample and a reagent to generate a measurement signal indicating whether an analyte is present, and if so, at what concentration. The post-analysis workstation may be adapted to perform any kind of post-processing on a sample, a sample container, and / or a sample container carrier.

[0088] Deadlock can be prevented. Deadlock is a situation that prevents a group of carriers from making the next move because other carriers already occupy the zone and cannot move either. If, for each zone, zone-to-zone (logical zone / plane orientation) movement is only possible in one X direction and one Y direction, then the minimum deadlock situation involves four carriers. In this case, the four carriers are located in an area of ​​2x2 zones. Each zone is occupied by one carrier. At each zone in the area of ​​2x2 zones, it is allowed and expected to move clockwise or counterclockwise to the next zone in the area of ​​2x2 zones. However, due to occupation, such movement is impossible because the possibility of each carrier moving depends on the possibility of another carrier moving. Therefore, there is a circular dependency of carriers and their next move. There may also be larger deadlock situations caused by the same mechanism but involving more carriers.

[0089] If for each zone, zone-to-zone (logical zone / plane position) movement is possible in each horizontal direction, the minimal deadlock case involves two vehicles. In this case, similar to the 2x2 deadlock, the two vehicles are located in the area of ​​2x1 (1x2) zones. Vehicles involved in a larger deadlock may be placed in all transport positions of a rectangular (nxm) area of ​​the transport plane. In addition to this case, deadlocks may also involve arbitrary closed sequences of plane positions, each of which contains a vehicle that cannot be moved due to cyclic dependencies. Deadlocks may even involve plane positions of different planes and ramps / lifts. The embodiments presented here apply similarly.

[0090] From each plane location / logical zone, the carrier may typically be able to move in two X directions and two Y directions (left / right and up / down). Alternatively, the carrier may be restricted to move in only one X direction and in one Y direction from each plane location / logical zone. However, the allowed directions of movement do not have to be the same for each plane location / logical zone. For example, at a first plane location / logical zone, the carrier may only move upward and to the right, and at a second plane location / logical zone, the carrier may only move downward and to the left. Restrictions may be provided to prevent 2x1 (1x2) deadlock.

[0091] A specific pattern for position-to-position movement allowed on the transport plane can be generated, for example to avoid 2x2 deadlock. At each plane orientation, the carrier can only move in one X direction and one Y direction. The plane can be rectangular (or composed of rectangular parts). The plane can be divided into a secondary region of nxn plane orientation (or divided into a rectangular region of nxm plane orientation). Each region of the nxn plane orientation can correspond to a transport module. The carrier can be allowed to move in the first X direction in the upper part (half) of each of the secondary regions of the transport plane, and in the lower part (half) of each of the secondary regions, in a second X direction different from the first X direction, i.e., the first and second X directions lead to opposite directions. The carrier can be allowed to move in the first Y direction in the left part (half) of each of the secondary regions of the transport plane, and in the second Y direction different from the first Y direction in the right part (half) of each of the secondary regions, i.e., the first and second Y directions lead to opposite directions. Each secondary region can include a region of the 2x2 plane orientation in which circular movement is allowed. In addition, the area of ​​the 2x2 plane orientation included by four such secondary areas can allow circular movement. In these areas of the 2x2 plane orientation, four allowed movements can define allowed circular movements (which cause deadlock in each area of ​​this area of ​​the 2x2 plane orientation). One of the four allowed movements can be removed (to prevent deadlock situations). Alternatively, more than one, preferably two of the four allowed movements can be removed. Preferably, one or more allowed movements removed are movements that are not expected to be frequently used. In particular, one or more allowed movements removed are movements perpendicular to the main transport flow.

[0092] There are two ways to obtain routes where 2x2 deadlock is avoided: (1) define the allowed directions for determining routes according to the specific pattern described above and / or (2) when computing the optimized set of offline routes, add criteria to avoid any subset of zone-to-zone moves involved in the set of routes to define a region of 2x2 plane positions where circular moves are allowed.

[0093] For all identified deadlocks, the size (the number of positions that need to be occupied to form a deadlock) can be calculated. Smaller deadlocks, such as 2x2 deadlocks, may occur more often than larger deadlocks. In addition, larger deadlocks may contain smaller deadlocks. Avoiding smaller deadlocks and larger deadlocks may include reducing the solution space used to find optimized routes. Priority may be given to eliminating opportunities to generate smaller deadlocks. A deadlock size threshold may be provided. Calculating an optimized set of offline routes may avoid routes that may result in deadlocks whose size is less than or equal to the deadlock size threshold.

[0094] Deadlock can be prevented by changing the rules for vehicles. Each vehicle can reserve a corresponding track segment. The first vehicle can reserve a first track segment. The first track segment can include one or more subsequent plane positions on the route of the first vehicle starting from the current plane position of the vehicle. The position of the reserved first track segment may be blocked by other vehicles. The first plane position after the reserved first track segment on the route of the first vehicle can be marked. The marked plane position may not be blocked by other vehicles. The second vehicle may not reserve a second track segment ending with a marked plane position, that is, the last plane position of the second track segment of the second vehicle starting from the current position of the second vehicle may not be a marked plane position. The track segment can include a fixed number of plane positions, such as one, two or more. Alternatively, different vehicles can reserve track segments with different numbers of plane positions.

[0095] The method may further include detecting a fracture (and / or unusable) plane position including at least one fracture plane orientation, wherein the at least one fracture plane orientation is included in at least one of the determined offline routes, and dividing at least one of the determined offline routes into a first sub-route and into a second sub-route. In response to determining the fracture plane position, a region including at least one fracture plane orientation, at least one working plane orientation corresponding to the first sub-route, and at least one working plane orientation corresponding to the second sub-route may be determined. The method may further include calculating an optimized connection route between at least one working plane orientation corresponding to the first sub-route and at least one working plane orientation corresponding to the second sub-route, the calculation comprising solving an optimization problem in which the connection route between at least one working plane orientation corresponding to the first sub-route and at least one working plane orientation corresponding to the second sub-route is optimized while excluding at least one fracture plane orientation. The optimization may take into account a provided (existing) set of offline routes. The first sub-route, the connection route, and the second sub-route may replace at least one of the determined offline routes. The method may further include controlling the drive means so as to move the carrier along a predetermined off-line route on the transport plane, wherein at least one of the determined off-line routes is replaced by a first sub-route, a connecting route, and a second sub-route. Calculating the connecting route may be done at mid-run. Once a broken (or unusable) plane position is detected, the system may be stopped. After calculating the connecting route, the system may be started again.

[0096] For each of the determined offline routes, the maximum expected load / maximum expected number of carriers to be transported along (on) the route may be assigned. The method may include, during operation (at runtime), avoiding exceeding the maximum load / maximum number of carriers for one of the determined offline routes. For example, the method may include stopping the supply of additional carriers to a route that has reached its maximum load / its maximum number of carriers. When stopping the supply of additional carriers to a route, the additional carriers (or corresponding samples / racks) expected to be carried along the route may be temporarily buffered on or near the corresponding laboratory device until the load / number of carriers of the route is sufficiently low, i.e. the load / number of carriers of the route has dropped below its maximum load / its maximum number of carriers or below a predefined threshold assigned to the route. Alternatively or additionally, the method may include, in order to avoid exceeding the maximum load / maximum number of carriers for one of the determined offline routes, splitting the route that has reached its maximum load / its maximum number of carriers into two or more different offline routes.

[0097] For a laboratory sample dispensing system and / or a laboratory automation system, the above described embodiments in combination with a method of operating a laboratory sample dispensing system may be provided accordingly. DETAILED DESCRIPTION

[0098] Next, the embodiments are described with reference to the accompanying drawings by way of example. The following figures show:

[0099] Figure 1 Graphical representation of laboratory sample distribution system;

[0100] Figure 2 a graphical representation of a flow chart of a method of operating a laboratory sample dispensing system;

[0101] Figure 3A Graphical representation of an example of minimal deadlock;

[0102] Figure 3B A graphical representation of examples of allowed movements of a vehicle at a given planar location / logical zone;

[0103] Figure 3C A graphical representation of examples of disallowed vehicle movements at a given plane location / logical zone;

[0104] Figure 3D A graphical representation of the specific patterns of movement permitted on the transport plane;

[0105] Figure 4 A graphical representation of the reserved track segments and subsequently labeled plane positions / logic zones for the first vehicle;

[0106] Figure 5 a graphical representation of the split route;

[0107] Figure 6 The first situation that may occur when solving an optimization problem;

[0108] Figure 7 A second situation may occur when solving an optimization problem; and

[0109] Figure 8 A third situation may arise when solving an optimization problem.

[0110] Figure 1 A diagrammatic representation of a laboratory sample distribution system is shown. The laboratory sample distribution system comprises: a plurality of carriers 4 configured to carry one or more sample containers containing samples to be analyzed by a laboratory device 3; a transport plane 1 assigned to the laboratory device 3 and providing support for the sample container carriers 4; and a drive device 13 configured to move the plurality of carriers 4 between planar orientations 5 provided on the transport plane 1 in response to a drive control signal.

[0111] The transport plane 1 may include a plurality of transport modules 12. The transport plane 1 may include a plurality of plane locations / logical zones 5. In the illustrated case, one plane location / logical zone 5 defines one plane position 5'. Each transport module 12 may be assigned to a respective plane location / logical zone 5. Each laboratory device 3 may be assigned to one or more specific plane locations / logical zones 5. Figure 1 Also shown is a directed graph model 8 of the transport plane 1. The directed graph model 8 includes a plurality of nodes 9 and a plurality of arcs 10 connecting the nodes 9. Each of the nodes 9 of the directed graph model 8 may correspond to a corresponding plane position 5'. Alternatively, for a subset of the nodes 9, each node 9 may correspond to a corresponding plane position 5'. Each of the arcs 9 of the directed graph model 8 may correspond to a movement between two corresponding plane positions 5'. Alternatively, for a subset of the arcs 9, each arc 9 may correspond to a movement between two corresponding plane positions 5'.

[0112] A certain laboratory device 3 may correspond to certain plane orientations 5 / plane positions 5'. One of these plane positions 5' may be a transfer position 16. Each laboratory device 3 may be assigned to one or two transfer positions 16. A carrier 4 located at a first transfer position 16 assigned to a specific laboratory device may be transferred from the transfer position 16 to the specific device. Alternatively, if a carrier 4 with a sample is located at the first transfer position 16 assigned to the specific laboratory device 3, the sample may be transferred from the transfer position 16 to the specific device 3. A carrier 4 located in a specific device 3 may be transferred to a second transfer position 16 assigned to the specific laboratory device 3. Alternatively, if a carrier 4 without a sample is located at the second transfer position 16 assigned to the specific laboratory device 3, the sample may be transferred from the specific device 3 to the carrier 4 at the second transfer position 16. The first and second transfer positions 16 may be assigned to the same specific laboratory device 3. The first and second transfer positions 16 may correspond to the input and output (input plane position / orientation and output plane position / orientation) of the laboratory device 3. The first and second transfer positions 16 may correspond to the same or different planar positions 5' / planar locations 5 assigned to a specific laboratory device 3. It should be noted that only one carrier 4 may be provided on one planar location 5.

[0113] Via the transport plane 1, the vehicle 4 is moved between different plane positions 5'. In particular, the vehicle 4 can be moved between pairs of plane positions 11. The movement can correspond to a route 6 for the vehicle 4. The first plane position 5' of the route can be a starting plane position, while the last plane position 5' of the route can be a destination plane position, or vice versa. The starting plane position and the destination plane position can correspond to end point positions and / or pairs of plane positions 11. For each pair of plane positions 11, a different route 6 can be provided. Each plane position pair 11 may include two plane positions 5, such as a starting plane position 11' and a destination plane position 11", i.e., two endpoint positions. First frequent endpoint position data may be determined, indicating a first selection 14 of plane positions that most frequently provide endpoints of the travel route 6 for the vehicle 4. This first selection 14 of plane positions may correspond to a first group of plane position pairs 11. The first group of plane position pairs 11 may correspond to transfer positions 16, in particular transfer positions 16 that are frequently visited by the vehicle. Second frequent endpoint position data may be determined, indicating a first selection 14 of plane positions that less frequently provide endpoints of the travel route 6 for the vehicle 4. The second selection 15 of plane positions may correspond to a second set of plane area pairs 11. The second selection 15 may not correspond to transfer positions 16. The second set of plane position pairs 11 may not correspond to transfer positions 16. The second set of pairs may include pairs 11 that do not correspond to the first selection 14 and / or pairs 11 in which one position in each pair 11 corresponds to the first selection 14 and the other position in each pair 11 corresponds to the second selection 15. The second set of plane position pairs 11 may correspond to transfer positions 16 that are visited less frequently by the vehicle 4. An alternating sequence of nodes 9 and arcs 10 may form a route 6 in which the first and last elements are nodes 9 corresponding to plane position pairs 11.

[0114] Figure 1 Four endpoints are shown. The four endpoints may correspond to a selection of plane positions 14, 15. There are more than twelve possible connection routes between the four endpoints. However, for simplicity, Figure 1 Only two routes are shown. Each endpoint can form a plane position pair 11 with any other endpoint. Accordingly, these four endpoints can define twelve different plane position pairs 11 (start / end pairs). In general, m endpoints can define 2·b(m,2)=m·(m-1) different plane position pairs 11 (here, b() is the binomial coefficient). However, according to Figure 1, the two first endpoints 11', 11" correspond to a first selection of planar positions 14, and the two other endpoints (second endpoints) correspond to a second selection of planar positions 15. The first endpoints can define a first pair of planar positions. The second endpoints can define a second pair of planar positions. The first flow corresponding to the first pair of planar positions can be higher than the second flow corresponding to the second pair of planar positions. Therefore, the number of vehicles 4 traveling between the first pair of planar positions can be higher than the number of vehicles 4 traveling between the second pair of planar positions, for example, within a given time interval.

[0115] Figure 2 A graphical representation of a flow chart of a method of operating a laboratory sample distribution system is shown. Figure 2 The method comprises: before moving the carrier 4 on the transport plane 1, predetermining 20 an offline route on the transport plane by one or more processors of the data processing device (for example based on the transfer position). Predetermining 20 comprises: determining 21 a model representing a transport plane having plane positions and position-to-position movements between the plane positions associated with a plurality of carriers. Further, predetermining 20 comprises: using the model to calculate a set of optimized offline routes between pairs of plane positions from the plurality of plane positions, the calculation comprising solving 22 an optimization problem in which the routes between the pairs of plane positions are simultaneously optimized with respect to each other. Predetermining 20 further comprises: providing 23 the optimized set of offline routes as offline routes on the transport plane. The method further comprises: controlling the drive device 30 so as to move the carrier 4 along the predetermined offline route 6 on the transport plane 1.

[0116] The plane position pairs 11 may include a first and a second group of plane position pairs 11. The predetermination 20 may include two runs, namely a first and a second predetermination. In the first predetermination, the route 6 for the first group of plane position pairs 11 may be determined, and in the second predetermination, the route 6 for the second group of plane position pairs 11 may be determined. Alternatively, calculating 22 the optimized set of offline routes may include two runs, one run for the first group of plane position pairs 11 and one run for the second group of plane position pairs. In each case, priority may be given to the first run. The first run may be performed before the second run. During the second run, the route determined via the first run may be fixed. Figure 1 In the example, the first route 6 ′ may be calculated via a first run and the second route 6 ″ may be calculated via a second run.

[0117] For example, calculating 22 the optimized set of offline routes may include a first run for determining the first route 6 and, independently of the first run, calculating 22 the optimized set of offline routes may include a second run for determining the second route 6. Subsequently, the first route and the second route 6 may be re-optimized simultaneously based on the first route and the second route 6 and / or based on (all) planar position pairs 11. This subsequent step may correspond to determining 23 an optimized set of offline routes from a plurality of routes.

[0118] Figure 3A A graphical representation of an example of a minimum deadlock in the case where zone-to-zone (logical zone / plane orientation) movement can be performed in only one X direction and one Y direction for each zone is shown. In this case, the four plane orientations 5 of the area of ​​2x2 plane orientations 5 are occupied by four vehicles 4. Therefore, each of the four plane orientations 5 is occupied by one vehicle 4. For each plane orientation 5, for the area of ​​2x2 plane orientations 5, a counterclockwise movement 31 to another plane orientation of the area of ​​2x2 plane orientations 5 is allowed. Further, for each of the four vehicles 4, such movement is also expected. However, since each of the four plane orientations 5 is occupied, none of the four vehicles 4 can reserve a subsequent plane orientation 5 and / or can perform the expected movement. This produces a deadlock. Similarly, although not shown in the figure, a deadlock can also be formed by more than four vehicles 4 and plane orientations 5. In each such case, the deadlocked vehicles 4 can form a closed circle. An adjacency matrix corresponding to the blocked vehicles and an arc corresponding to the next movement 4 can be provided. By calculating the (finite) power of the adjacency matrix, it can be determined whether the blocked vehicles 4 form a circle, ie whether the blocked vehicles form a deadlock.

[0119] Figure 3B A graphical representation of an example of allowed movement 32 of a vehicle 4 at a given planar orientation 5 is shown. At a given planar orientation 5, the vehicle 4 may be allowed to move in only one X direction and one Y direction. Figure 3B In FIG. 4 , an embodiment is shown which allows the carrier 4 to move leftward in the X direction and upward in the Y direction. Figure 3C Additionally, impermissible vehicle movements 33 are shown for a given plane position 5 .

[0120] Figure 3D A graphical representation of a specific pattern of position-to-position movements 32 allowed on the transport plane 1 is shown. Figure 3D, the transport plane 1 is divided into secondary areas 35 of 6x6 planar orientations. Each area 35 of the 6x6 planar orientation corresponds to a transport module 12. In the upper half of each secondary area 35, the carrier 4 is allowed to move to the left in the X direction, while in the lower half of each secondary area 35, the carrier 4 is allowed to move to the right (only) in the X direction. In the left half of each secondary area 35, the carrier 4 is allowed to move downward in the Y direction, while in the right half of each secondary area 35, the carrier 4 is allowed to move upward (only) in the Y direction. In this model, each secondary area 35 includes an area 34 of a 2x2 planar orientation located in the center, which may produce a deadlock. In this area 34 of the 2x2 planar orientation, clockwise carrier movement is allowed. In addition, the model includes additional such areas 34 of 2x2 planar orientations, wherein each planar orientation 5 of each of the additional such areas 34 of the 2x2 planar orientation corresponds to a corner orientation 5 of the secondary area 35. In Figure 3D In FIG. 1 , each such region 34 of 2x2 plane orientations (which may form a deadlock) is marked with an exclamation mark. The four plane orientations involved in these 2x2 plane orientations come from 4 adjacent corners of 4 regions. There is one 6x6 region that is fully represented, and 8 parts of the 6x6 region (only partially represented in FIG. 1 ). Figure 3D middle).

[0121] Figure 4 A graphical representation of a reserved track segment 42 and a subsequently marked planar position 43 of a first vehicle 4 ′ is shown. Figure 4 A method for preventing deadlock is shown. A track segment 42 of a first vehicle 4' reserves a plane orientation 5 along its route 41. The first plane orientation 5 of the segment 42 corresponds to the plane orientation 5 that the first vehicle 4' will enter next starting from its current plane orientation 5. Starting from the first plane orientation 5, the track segment 42 includes one or more immediately adjacent additional plane orientations 5 until the last plane orientation 5 of the track segment 42. Figure 4 , the track segment 42 includes two plane orientations 5, namely the first and the last. The plane orientations 5 of the track segment 42 of the first vehicle 4' may be reserved for the first vehicle 4', i.e., these plane orientations 5 may be blocked for vehicles 4 other than the first vehicle 4'. The plane orientations 5 immediately after the last zone of the segment 42 of the first vehicle 4' on the route 41 of the first vehicle 4' may be marked for the first vehicle 4'. The marked plane orientations 43 of the first vehicle 4' may not be reserved by other vehicles 4, i.e., the marked plane orientations 43 may not be included in the segments 42 of other vehicles 4. In Figure 4 In particular, the fourth carrier 4″ is therefore not allowed to reserve the next two plane positions to the right of its current position because the fourth carrier is not allowed to reserve the marked plane position 43.

[0122] Figure 5 A graphical representation of a split route 51 is shown. If the traffic of the determined route 51 has a high flow rate, for example if the flow rate exceeds a threshold, the route 51 may be divided into several sub-routes 52, 53. Figure 5 In the case of a high traffic volume, the route 51 with high traffic volume is divided into two sub-routes 52, 53 (to reduce the possibility of congestion at the intersection with another route). The sub-routes 52, 53 include overlapping sections, separation / merging sections, and parallel sections. The traffic volume of the route 51 with high traffic volume can be divided 50% to the upper route 52 and 50% to the lower route 53. Alternatively, the traffic volume of the route 51 with high traffic volume can be allocated to the sub-routes 52, 53 according to different weights. For example, the shortest of the sub-routes 52, 53 of the split route 51 can be set with the highest percentage of traffic volume. In order to avoid collisions at the intersection 54 of the routes, the vehicle 4 assigned to the split route 51 can alternatively follow one of the sub-routes 52, 53.

[0123] Figures 6 to 8 Three situations that may occur when solving the optimization problem are shown. These three situations refer to three possibilities for providing two routes connecting two pairs of plane positions. Figure 6 In the first case, the first route 61 connects the first pair of plane positions, and the second route 62 connects the second pair of plane positions. Figure 7 In the second case, the additional first route 71 connects the first plane position pair, and the additional second route 72 connects the second plane position pair. Figure 8 In the third case, the additional first route 81 connects the first planar position pair, and the additional second route 82 connects the second planar position pair.

[0124] The first route 61, 71, 81 may correspond to a higher flow rate than the second route 62, 72, 82. The penalty for the corresponding route may be proportional to the flow intensity of the corresponding route. According to this example, during the solution of the optimization problem, the following penalty strategy (constraint) may be applied. For routes with low flow intensity, the following applies: (i) route length: 5 penalty points for each zone (plane position or plane orientation) used in the route; (ii) route junction: 50 penalty points for a route that joins with another route; (iii) route curve: 2 penalty points for each curve in the route. For routes with high flow intensity, the following applies: (i) route length: 10 penalty points for each zone (plane position or plane orientation) used in the route; (ii) route junction: 100 penalty points for a route that joins with another route; (iii) route curve: 4 penalty points for each curve in the route. In the first case, the route with higher flow includes 10 zones, 0 intersections, and 2 curves. The route with lower flow includes 8 zones, 0 intersections, and 5 curves. Therefore, according to the first case, the route with higher traffic is associated with 108 penalty points, while the route with lower traffic is associated with 50 penalty points. Therefore, the first case is associated with 158 penalty points. In the second case, the route with higher traffic includes 10 areas, 2 intersections and 2 curves. The route with lower traffic includes 8 areas, 2 intersections and 1 curve. Therefore, according to the second case, the route with higher traffic is associated with 308 penalty points, while the route with lower traffic is associated with 142 penalty points. Therefore, the second case is associated with 450 penalty points. The second case is associated with more penalty points than the first case. In the third case, the route with higher traffic includes 10 areas, 0 intersections and 1 curve. The route with lower traffic includes 8 areas, 0 intersections and 1 curve. Therefore, according to the third case, the route with higher traffic is associated with 104 penalty points, while the route with lower traffic is associated with 42 penalty points. Therefore, the third case is associated with 148 penalty points. Therefore, when solving the optimization problem, the first case is preferred to the second case, and the third case is preferred to the first and second cases.

Claims

1. A method of operating a laboratory sample distribution system, wherein the laboratory sample distribution system comprises: - a plurality of carriers (4) configured to carry one or more sample containers containing samples to be analyzed by the laboratory device (3); - a transport plane (1) which is assigned to the laboratory device (3) and provides support for the plurality of carriers (4); as well as - a drive device (13) configured to move the plurality of carriers (4) between planar positions (5) provided on the transport plane (1) in response to a drive control signal; The method comprises: - before moving the vehicle (4) on the transport plane (1), one or more processors of a data processing device predetermine an offline route (6) on the transport plane (1), the predetermination comprising: - determining a model representing the transport plane (1) having plane positions (5') and position-to-position movements between the plane positions (5') associated with the plurality of carriers (4) - using the model to calculate an optimized set of off-line routes between pairs of plane positions from a plurality of plane positions (5'), the calculation comprising solving an optimization problem in which the routes between the pairs of plane positions are optimized simultaneously; and - providing the optimized set of off-line routes as off-line routes (6) on the transport plane (1); and - controlling the drive device (13) so as to move the carrier (4) along a predetermined off-line route (6) on the transport plane (1).

2. A method according to claim 1, wherein the model is a directed graphical model (8) of the transport plane (1), wherein the nodes (9) of the directed graphical model (8) are assigned plane positions (5'), and the arcs (10) connecting the nodes (9) of the directed graphical model (8) are assigned position-to-position movements between two plane positions (5').

3. The method according to claim 1 or 2, wherein the optimization problem is one of the following: -Multi-object network flow problems, especially multi-object network flow problems in directed graphs; - the shortest path problem; and -Minimum flow problem.

4. The method according to at least one of the preceding claims, wherein the optimization problem is solved by applying a MIP solver.

5. The method according to claim 2, further comprising: In the data processing device, - providing first frequent endpoint position data, the first frequent endpoint position data indicating a first selection (14) of plan positions of endpoints that most frequently provide a travel route for the vehicle (4); as well as - Determining the directed graphical model (8) of the transport plane (1), wherein a first node (9) of the directed graphical model (8) is assigned a plane position (5') from a first selection (14) of plane positions, and a first arc (10) starting and / or ending with the first node (9) of the directed graphical model (8) is assigned a position-to-position movement from and / or to a plane position (5') from the first selection (14) of plane positions.

6. The method according to claim 2 or 5, further comprising: In the data processing device, - providing second frequent endpoint position data indicating a second selection of planar positions (15) that less frequently provide endpoints of a travel route (6) for the vehicle (4), wherein the second selection of planar positions (15) is different from the first selection of planar positions (14); and - Determining the directed graphical model (8) of the transport plane (1), wherein a second node (9) of the directed graphical model (8) is assigned a plane position (5') from a second selection (15) of plane positions, and a second arc (10) starting and / or ending at the second node (9) of the directed graphical model (8) is assigned a position-to-position movement from and / or to a plane position (5') from the second selection (15) of plane positions.

7. The method according to at least one of the preceding claims, further comprising: In the data processing device, - providing flow data indicating a predicted number of vehicles (4) travelling between the pair of planar positions (11) within a time interval; as well as - calculating the optimized set of off-line routes between pairs of plane positions from the plurality of plane positions (5'), the calculation being performed based on a predicted number of the vehicles (4) travelling between pairs of plane positions (11) within the time interval.

8. The method of claim 7, wherein providing traffic data further comprises at least one of: - Provide flow data determined from a sample order list; - providing flow data determined from historical data indicative of historical operation of the laboratory sample dispensing system; - providing flow data determined from workflow data indicating a workflow for the one or more sample containers to be carried by the carrier (4); - providing flow data determined from the measured current and / or recent number of vehicles (4) transported; as well as - Provides flow data determined from simulations.

9. The method according to at least one of the preceding claims, the controlling of the drive device (13) further comprising: - in the driving device (13), receiving a reservation request from a vehicle (4), the vehicle travelling on a selected off-line route (6) from the predetermined off-line routes (6) and being located at a current route position along the selected off-line route (6), the reservation request indicating a request for reserving a subsequent route position along the selected off-line route (6); - verifying by the driving device (13) whether the subsequent route position is free for travel; as well as - if the drive device (13) verifies that the subsequent route position is free for travel, the vehicle (4) is moved along the selected off-line route (6) from the current route position to the subsequent route position.

10. The method according to at least one of the preceding claims, wherein the calculation of the optimized set of off-line routes between pairs of plane positions from the plurality of plane positions (5') via solving the optimization problem further comprises applying at least one constraint selected from the group consisting of: - minimizing the route length of each of the offline routes; - minimizing a weighted route length for each of said offline routes; - minimizing the number of route curves for each of said offline routes; - minimizing the number of offline routes that join and / or intersect with another offline route; -Even distribution of vehicle flow at each plane position (5'); - limiting the position-to-position movement between two planar positions (5') to movement only between adjacent planar positions; - Exclude the plane position (5') reserved for the vehicle queue; - evenly distributing the predicted wear of the planar positions over the planar positions (5') of the transport plane (1); - minimizing the energy consumption of the laboratory sample distribution system; as well as - Minimize / avoid regions of 2x2 plane orientations with four intersections.

11. The method according to at least one of the preceding claims, wherein the predetermination (20) of an offline route further comprises: In the data processing device, - receiving first route traffic information indicating high vehicle traffic for a first offline route (51); as well as - Splitting the first off-line route (51) into two or more different off-line routes (52, 53).

12. The method according to at least one of the preceding claims, wherein the predetermination (20) of an offline route further comprises: In the data processing device, - receiving second route traffic information indicating high vehicle traffic for a second offline route; as well as - when determining a plurality of offline routes and / or determining an optimized set of offline routes, preventing a route adjustment of the second offline route.

13. The method according to at least one of the preceding claims, wherein the calculation of the optimized set of off-line routes between pairs of plane positions from the plurality of plane positions (5') using the model further comprises: In the data processing device, - receiving first vehicle traffic information indicating a first vehicle traffic scenario for the plurality of offline routes (6); - determining a first plurality of off-line routes (6) between said pairs of planar positions (11) from said plurality of planar positions (5'); - receiving second vehicle traffic information indicating a second vehicle traffic scenario for the plurality of offline routes (6), wherein the second vehicle traffic scenario is different from the first vehicle traffic scenario; as well as - determining a second plurality of off-line routes (6) between said pairs of planar locations (11) from said plurality of planar locations (5').

14. The method according to at least one of the preceding claims, the controlling of the drive device (13) further comprising: - operating said laboratory sample dispensing system at runtime; as well as - If it is determined that a runtime route cannot be determined for the vehicle (4) at runtime, selecting an offline route (6) from the optimized set of offline routes (6).

15. The method according to at least one of the preceding claims, wherein: - said predetermination of an offline route further comprises: - determining a first set of optimized off-line routes (6); - assigning first application parameters to the optimized first set of offline routes (6); - determining an optimized second set of off-line routes (6) different from the optimized first set of off-line routes (6); and - assigning second application parameters to the optimized second set of offline routes (6); and - The control of the drive device further comprises: - receiving application information indicating current application parameters; and - If it is determined that the current application parameter matches the first application parameter or the second application parameter, one of the optimized first set of offline routes and the optimized second set of offline routes is selected to control the drive device (13).

16. A laboratory sample distribution system, comprising: - a plurality of carriers (4) configured to carry one or more sample containers containing samples to be analyzed by the laboratory station (3); - a transport plane (1) which is assigned to the laboratory device (3) and provides support for the plurality of carriers (4); as well as - a drive device (13) configured to move the plurality of carriers (4) between planar positions (5) provided on the transport plane (1) in response to a drive control signal; and Configured to: - before moving the vehicle on the transport plane, one or more processors of a data processing device predetermine an off-line route (6) on the transport plane (1), comprising: - determining a model representing the transport plane (1) having plane positions (5') and position-to-position movements between the plane positions (5') associated with the plurality of carriers (4); - using the model to calculate an optimized set of off-line routes between pairs of plane positions from a plurality of plane positions (5'), the calculation comprising solving an optimization problem in which the routes between the pairs of plane positions are optimized simultaneously; and - providing the optimized set of off-line routes as off-line routes (6) on the transport plane (1); and - controlling the drive device (13) so as to move the carrier (4) along a predetermined off-line route (6) on the transport plane (1).

17. A laboratory automation system comprising the laboratory sample distribution system according to claim 16 and a plurality of laboratory devices (3).

18. The laboratory automation system according to claim 17, wherein the plurality of laboratory devices (3) include one or more laboratory devices selected from: a laboratory device for pre-analysis; a laboratory device for sample analysis; and a laboratory device for post-analysis.

Images