ITTO980233A1 - LOAD BALANCING METHOD IN AN OBJECT SEQUENCING PROCESS, IN PARTICULAR POSTAL OBJECTS. - Google Patents

Description

DESCRIPTION

of the patent for industrial invention

The present invention relates to a load balancing method in an object sequencing process.

In particular, the present invention finds an advantageous, but not exclusive, application in load balancing in a sequencing process of postal objects, to which the following discussion will make explicit reference without thereby losing generality.

In fact, the method according to the present invention can also be used for load balancing in any sequencing process carried out by means of one or more devices capable of separating flows of objects to be arranged in a predetermined sequence for addressing to delivery locations. or destination, where the capacity of the aforementioned device or devices is limited or expensive.

Machines for postal sorting and sequencing are known which receive a flow of postal objects arranged in a completely random manner at their input and generate a sequenced flow of postal objects at the output, i.e. a flow of postal objects arranged according to a predetermined progressive order that allows sequential distribution by one or more mobile postmen along a predefined path.

The progressive order in which the postal objects are arranged in the exits from the postal machine is generically defined by a sequence of adjacent delivery or destination locations corresponding to house numbers, or groups of house numbers, of buildings arranged along the route where they must be delivered such postal items.

Known postal machines generally comprise an input, also called induction, receiving a set of postal objects, a plurality of outputs to which respective containers are associated in which groups of postal objects are arranged and a transport and addressing device (SORTER) interposed between the entry and exit is controlled by an electronic processing unit having the purpose of directing each postal object towards a respective exit on the basis of a code generally stamped on the postal object itself.

Furthermore, each postman carrying out the delivery service of postal objects is assigned a respective group of outlets of the postal machine distinct from that assigned to another postman and from which at the end of the sequencing process he will extract the postal objects entrusted to him for delivery.

The sequencing process carried out by the postal machine generally provides for the execution of a plurality of sorting cycles, temporally consecutive, by which groups of postal objects already subjected to previous sorting are reintroduced into the entrance of the postal machine and addressed to outputs associated with containers from which postal items deposited in a previous sorting cycle have been removed.

At the end of these sorting cycles, groups of postal objects are extracted from the postal machine, arranged in a predetermined progressive order that allows them to be sequentially distributed by a postman moving along a subsection of a predefined path.

There are also postal machines which have two or more inputs communicating with a single transport and addressing device which sends the postal objects towards groups of outputs associated with respective inputs.

Each entrance of the postal machine cannot use outputs in common with the other inputs and in this way, although the transport and addressing device is shared, such postal machines actually comprise several independent and separate machines.

A common drawback of known sequencing processes lies in the fact that these processes provide that each exit of the postal machine is associated with a predetermined number of adjacent delivery locations that the postman assigned to this exit will have to serve and do not take into account the quantity of postal objects. which must be delivered to the delivery points.

As a result, a situation of filling an exit of the postal machine may occur during the execution of a sequencing process, for example because a large user receiving a large quantity is located in one of the delivery locations associated with that exit. of postal items.

In such cases, the interruption of the current sorting cycle is therefore inevitable in order to extract the postal items collected in the filled exit.

Therefore, the situation of filling an exit of the postal machine is enormously undesirable as it involves considerable waste of time due to both the interruption of the sorting cycle and the numerous precautions that must be adopted in the management of postal objects before resuming the sequencing process. same.

This situation becomes particularly burdensome when the filling of an output of the postal machine is repeated several times during the various sorting cycles carried out during a work shift.

The object of the present invention is that of realizing a load balancing method in a sequencing process of postal objects which allows to avoid as much as possible situations of filling the outlets of the postal machine which carries out this process.

According to the present invention, a load balancing method is implemented in a sequencing process of objects, in particular postal objects, said sequencing process comprising the execution of a first and at least a second consecutive sorting cycle and being carried out by means of a machine receiving in input a set of objects and supplying said objects identified and separated on its outputs according to specific sorting rules; in a given sorting cycle, the objects being supplied to the machine outlets according to a respective predetermined sorting criterion and being neatly supplied again to the machine itself for the execution of a subsequent sorting cycle; the machine outputs being divided, in the last sorting cycle of the sequencing process, into distinct subsets and a plurality of respective places of destination being assigned to each subset according to an assignment criterion correlated to the order of distribution of the objects in the places of destination ;

said balancing method being characterized in that it comprises the step of carrying out, in at least one of said sorting cycles, at least one of the following procedures:

a) a first load balancing procedure of the machine outputs at the end of the current sorting cycle based on at least one of the following balancing criteria:

• displacement of destination locations from the exits to which they are assigned to respective logically contiguous exits according to a first predetermined logical criterion and belonging to the same subset;

• assignment of further outputs to said subsets on the basis of a criticality order of the outputs already assigned to the subsets themselves; And

• bisection of places of destination into respective pairs of virtually distinct places of destination and assignment of each pair of virtually distinct places of destination to a respective pair of logically contiguous outputs according to a second predetermined logical criterion and belonging to the same subset; and

b) a second load balancing procedure of the machine outputs at the end of the current sorting cycle based on at least one of the following balancing criteria:

<•> distribution on the exits used in the current sorting cycle of destination locations assigned to each exit at the end of the next sorting cycle;

<■> displacement of destination locations from the exits to which they are assigned to respective logically contiguous exits according to a third predetermined logical criterion;

<•> exchange of destination places between logically contiguous outputs according to a fourth prefixed logical criterion; And

• bisection of destination places into respective pairs of virtually distinct destination places and assignment of each pair of virtually distinct destination places to a respective pair of logically contiguous exits according to a predetermined fifth logical criterion.

For a better understanding of the present invention, a preferred embodiment is now described, purely by way of non-limiting example and with reference to the attached drawings, in which:

Figure 1 schematically illustrates a postal machine for postal objects;

Figure 2 illustrates a matrix constituting a matrix representation of the load of postal objects in each output of a postal machine carrying out a sequencing process consisting of two sorting cycles; And

- Figures 3a-3k illustrate a flow diagram relating to the balancing method object of the present invention.

In Figure 1, the number 1 indicates, as a whole, a postal machine having an entrance I suitable for receiving a flow F of postal objects 2 (for example letters, postcards, enveloped documents or documents that are generally flat and of approximately rectangular shape) arranged in sequence (for example stacked) and conveyed towards the inlet I itself by means of a transport device 4 of a known type (for example of the belt type); and a plurality (N) of distinct outlets U1, U2, U3, ..., UN, each of which is conveniently associated with a removable container 6 (shown schematically) in which the postal objects 2 sent to this outlet are stacked.

The flow F of postal objects 2 comprises a plurality of postal objects on which a code, for example a bar code, suitable for identifying the place of delivery or destination of the postal object 2 itself has been imprinted in the previous step; these postal objects 2 are in any case arranged in a "disordered" type sequence, ie there is no progressive order nor any relationship between the disposition of the postal objects 2 and the progressive order according to which they must be subsequently distributed.

Associated with the inlet I of the postal machine 1 is a singling device 10 (schematically represented) receiving the postal objects 2 fed by the transport device 4 and able to extract postal objects 2 from the flow F and arrange each postal object 2 in a spaced position with respect to the other postal objects 2 of the flow F itself; a reading device 12 (schematically represented) receiving in input the postal objects 2 coming from the separator device 10 and able to read the code associated with the postal objects 2 themselves; a delay module 14 (schematically represented) receiving in input the postal objects 2 coming from the reading device 12; and a transport and addressing device 16 (SORTER) inside the postal machine 1 interposed between the output of the delay module 14 and the outputs U1, U2, UN.

The postal machine 1 is controlled by an electronic unit 22 of the programmable type, under the control of which the transport and addressing device 16 directs the flow F fed to the input I on all the N outputs of the postal machine 1, i.e. a common transport mode according to which each postal object 2 fed at input A can potentially be sent to any of the N outputs.

The displacements of a postal object along the transport and addressing device 16, i.e. the path T traveled by a postal object 2 along the transport device 16 from input I to an eneric output Ui, is determined by the code present on the object mail 2 itself read by the reading device 12.

In this regard, as will be described in more detail below, the electronic unit 22 is provided with a plurality of electronic tables (not shown) receiving in input (for example from the reading devices 12) the data associated with the code imprinted on each postal object 2 and supplying at the output a set of data identifying the selected exit Ui towards which said postal object 2 must be addressed.

The output data are then transmitted to the mail machine 1 to generate command signals for actuator members, for example paddle selectors, transmission members, etc. (not shown), which concur to create, along the transport device 16, that trajectory T which carries the postal object 2 towards the selected exit Ui.

The load balancing method according to the present invention will be described hereinafter with reference to a postal machine carrying out a sequencing process consisting of two temporally successive sorting cycles.

In particular, in the first sorting cycle, the postal objects are supplied to the input I of the postal machine 1 and then sorted to the outputs U of the postal machine 1 itself according to a first sorting criterion determined by the present balancing method. The postal objects are then neatly extracted from the exits U and re-inserted into the postal machine 1 itself through the entrance I according to a predetermined re-insertion order for the execution of the second sorting cycle.

In the second sorting cycle, the postal objects are sorted towards the outlets U according to a second sorting criterion determined by the present balancing method and then picked up from the outlets U themselves to be delivered to the respective delivery points by means of the postmen.

For descriptive convenience, the following description will refer to a postal machine 1 having ten outlets without thereby losing generality since the inventive principle underlying the balancing method object of the present invention can be applied without any modification to a postal machine having a any number of outputs.

The description of the balancing method according to the present invention will be carried out with reference to the matrix illustrated in Figure 2, which constitutes a matrix representation of the load of postal objects of each output U of the postal machine 1 at the end of each of the two sorting cycles constituting the sequencing process.

In particular, each row of the matrix represents the situation of the loading of postal objects of a respective exit U of the postal machine 1 at the end of the first sorting cycle, while each column of the matrix represents the situation of the loading of postal objects of a respective exit U of the postal machine 1 at the end of the second sorting cycle.

The situation of the load of postal objects of the outputs of the postal machine 1 at the end of the execution of this balancing method therefore allows to derive the sorting criteria to be used for the sequencing process.

As illustrated in Figure 2, the matrix has ten rows and ten columns indicated with respective progressive identification numbers. In particular, the identification numbers of the columns are arranged according to an increasing progressive order starting from the leftmost column of the matrix while the identification numbers of the rows are arranged according to an increasing progressive order starting from the lowest row of the matrix. .

Given the correspondence between the rows and columns of the matrix and the outputs of the postal machine 1 in the first and, respectively, in the second sorting cycle, each identification number of the rows and columns of the matrix of figure 2 also identifies a respective output U of the postal machine 1 at the end of the first and, respectively, of the second sequencing cycle.

However, the actual physical position of the outlets U in the postal machine 1 does not necessarily correspond to the progressive numbering of the rows and columns of the matrix of figure 2, i.e. the outlets of the postal machine 1 are not necessarily arranged according to an increasing progressive order corresponding to the order progressive increasing of the identification numbers of the rows and columns.

In other words, the exit of the postal machine 1 represented by the column identified by the number "1" may not physically be the first exit of the postal 1 itself, the output represented by the column identified by the number "2", a column which in the matrix is adjacent and subsequent to the first column, it may not physically be the second exit of the mail machine 1 itself, nor may it be adjacent and subsequent to the exit represented by the column identified by the number "1".

The same goes for the other rows and columns of the matrix in figure 2.

Therefore, the progressive numbering of the rows and columns is a "logical" type numbering, which corresponds to a "physical" arrangement of the outputs linked to it by a predetermined relationship stored in the electronic control unit 22 and used in the sequencing process to direct mail items to the desired exit.

In the remainder of the description, therefore, the term "logically contiguous outputs" will mean the outputs of the postal machine 1 represented by rows or columns identified with successive identification numbers, even though these outputs may be physically not adjacent to each other or their relative position. be deductible from the identification numbers of the respective rows or columns.

Furthermore, thanks to the one-to-one correspondence between the outputs of the postal machine 1 and the rows of the matrix in the first sorting cycle and between the outputs of the postal machine 1 and the columns of the matrix in the second sorting cycle, in the following of the description the term machine outputs mail 1 and the terms rows and columns of the matrix of figure 2 will be used indifferently in the two sorting cycles according to the descriptive convenience.

With reference again to the matrix of Figure 2, the boxes of the matrix also take on a precise meaning linked to the places of delivery or destination of the postal objects and in particular each box of the matrix defines a respective virtual position within the matrix to which it can be associated with a real address of a delivery location where postal items are to be delivered.

As the boxes of the matrix, as known, are uniquely identified by respective pairs of numerical values indicative of the row and column of the boxes themselves, each virtual position to which a delivery location can be associated can therefore be represented by means of the pair of identification numerical values of the row and column of the respective box.

Furthermore, given the one-to-one correspondence between the rows and columns of the matrix and the outputs of the postal machine 1 in the two sorting cycles, each pair of numerical values indicative of the row and column of the respective virtual position also represents the output of the postal machine 1 which postal items having a given place of delivery can occupy at the end of the first and, respectively, of the second sorting cycle.

In light of the above, the present balancing method is aimed at creating a table containing a plurality of data and defining a one-to-one relationship between all the possible codes imprinted on the postal objects 2 (and capable of identifying, as previously mentioned, respective delivery locations of the postal objects 2 themselves) and corresponding virtual locations associated with the delivery locations indicated by these codes and identified, each, by a pair of numerical values identifying the row and column of a respective box in the matrix.

From this matrix, the method will trivially deduce the sorting rules of the delivery locations on the outlets of the postal machine 1 at the end of the first and second sorting cycle, assigning a delivery location associated with a given box of the matrix respectively the output of the postal machine 1 corresponding to the row number of the box in the first sorting cycle and the output of postal machine 1 corresponding to the column number of the box in the second sorting cycle.

Furthermore, since the postal objects present in each exit of the postal machine 1 at the end of the second sorting cycle are arranged according to a predetermined order which allows their sequential distribution by a mobile postman along a predefined path, and since this order of distribution defined by a sequence of adjacent delivery locations corresponding to house numbers, or groups of house numbers, of buildings arranged along the path of the relative postman, the relationship between all the possible codes imprinted on the postal objects 2 and the corresponding virtual positions defined by the aforesaid table is such as to define a criterion for assigning the delivery locations to the respective exits of the postal machine 1 such as to allow compliance with the aforementioned order of distribution of the postal objects.

In detail, for each group of columns associated with the postmen, the relationship defined by the table assigns the delivery locations to the boxes in the matrix according to an increasing assignment criterion in the sense of the columns and rows as described below.

In particular, the delivery locations are assigned starting from the mailbox located in the first row of the first column of the matrix (mailbox located further down in the first column) up to the mailbox located in the last row of the first column of the matrix (topmost box in the first column) and then again starting from the box in the first row of the second column up to the box in the last row of the second column, and so on for the following columns.

The criterion with which the delivery locations are assigned to the virtual positions (i.e. to the boxes of the matrix) therefore defines a constraint on the boxes of the matrix of figure 2 prohibiting any exchange of positions between the numerical values represented in the boxes of a column.

With reference again to the matrix of Figure 2, the first and second numerical values indicative of the row and column of each box of the matrix of Figure 2 assume a precise meaning linked to the output Ui of the postal machine 1 that the postal object to be delivered 'in the place of delivery associated with the virtual position identified by this pair of values occupies at the end of the two sorting cycles of which the sequencing process is composed and in particular the first numerical value indicative of the line of the box identifies a Ui output of the postal machine 1 in which the postal object to be delivered to the place of delivery associated with the box will be placed at the end of the first sorting cycle while the second indicative numerical value of the box column identifies an exit Uj of the postal machine 1 in which the postal object to be deliver to the place of delivery associated with the box will be arranged at the end of the second or sorting cycle.

During each sorting cycle, once the code stamped on a postal object is identified, the virtual position associated with it is determined and the pair of numerical values identifying the row and the column defining this virtual position is used by the postal machine 1 to generate, through tables or sorting rules deduced directly from the matrix, control signals of actuator members, for example paddle selectors, transmission members, etc. (not shown) which concur to create, along the transport and addressing device 16, a trajectory T which carries the postal object 2 towards the selected exit Ui.

In particular, the postal machine 1 uses the first numerical value identifying the virtual position (line number) to determine the destination exit of the postal object at the end of the first sorting cycle and the second numerical value identifying the virtual position (number of the column) to determine the destination exit of the mail item at the end of the second sorting cycle.

With reference again to the matrix of Figure 2, numerical values are represented in the boxes of the matrix to which a place of delivery is assigned a particular meaning linked to the traffic of postal objects and in particular each numerical value can be correlated to the quantity of postal objects expected for delivery to the place of delivery associated with the box in which it is represented.

In particular, the numerical value represented in a box can be indicative of the quantity of postal objects in absolute or exact terms or in terms of expected traffic.

Therefore, a numeric value represented in a box - is indicative of the fact that this box has been assigned a delivery location having an expected or real traffic indicated by the numeric value itself while the boxes without numeric values represented in them represent available virtual positions for the assignment of the place of delivery.

Therefore, for descriptive convenience, in the following a box containing a numeric value will be indicated with the term "box occupied", meaning that a respective place of delivery is associated with it, while a box containing no numeric value will be indicated with the term " free box ", meaning that it is not associated with a respective place of delivery.

Furthermore, the sum of the numerical values of each row and each column is associated with a precise meaning linked to the load (i.e. the quantity of postal objects expected) at the output of the postal machine 1 corresponding to that row or column and in particular the sum of the values numbers of each row represents the load present on the output of the postal machine associated with that row at the end of the first sorting cycle while the sum of the numerical values of each column represents the load present on the output of the postal machine 1 associated with this column at the end of the second sorting cycle.

With reference again to the matrix of Figure 2, a first and a second vector are shown laterally and above the matrix itself, each containing a number of fields equal to the number of rows and, respectively, of the columns of the matrix, ie ten fields. Each of the fields of the first vector represented laterally to the matrix is arranged aligned to a respective row of the matrix itself and each of the fields of the second vector represented above the matrix is arranged aligned to a respective column of the matrix.

In each of the fields of the first vector a numerical value is represented equal to the sum of the numerical values represented in the boxes of the row to which it is aligned while in each of the fields of the second vector a numerical value is represented equal to the sum of the numerical values represented in the boxes of the column to which it is aligned.

In this way, the first and the second vector represent the total load of each output of the postal machine 1 at the end of the first and, respectively, of the second sorting cycle of which the sequencing process is composed.

As can be seen by analyzing the numerical values represented in the first and second vector above, the matrix representation of Figure 2 is indicative of a situation of loading of the outlets of the postal machine 1 which is very unbalanced between the first and second sorting cycle. In fact, the first exit (the one identified with number 1) must contain 290 postal items at the end of the first sorting cycle and only 84 at the end of the second cycle.

Therefore, by implementing the balancing method object of the present invention it will be possible to avoid situations of such strong imbalance which could cause, during the execution of the first sorting cycle, an overload of the first exit of the postal machine 1 and therefore could make it necessary to interrupt of the operation of the postal machine 1 itself in order to extract the postal objects in the outlets of which they have accumulated.

Figure 2 below the matrix also illustrates a third vector comprising four fields having different amplitudes, arranged aligned with respective "groups of columns of the matrix and identified with the letters A, B, C and D. In particular, each field of the third vector it is associated with the respective group of columns to which it is aligned and the identifying letter represented in it is indicative of the postman to which this group of columns is assigned.

In detail, according to what is illustrated in Figure 2, the field of the third vector identified with the letter A is associated with the first and second column of the matrix, the field identified with the letter B is associated with the third column, the field identified with the letter C is associated with the fourth, fifth, sixth and seventh columns and the field identified with the letter D is associated with the eighth, ninth and tenth columns.

Therefore, this type of representation illustrates the situation in which a first postman associated with the field identified with the letter A is assigned to the distribution of the postal objects arranged, at the end of the second sorting cycle, in the first and second exit of the postal machine 1, a second postman associated with the field identified with the letter B is assigned to the distribution of postal objects arranged in the third exit of the postal machine 1, a third postman associated with the field identified with the letter C is assigned to the distribution of postal objects arranged in the fourth, fifth, sixth and seventh exit of the postal machine 1, and a fourth postman associated with the field identified with the letter D is assigned to the distribution of the postal objects arranged in the eighth, ninth and tenth exit of the postal machine 1.

As previously said about the matrix representation of figure 2, it is evident how this matrix representation allows to fully and clearly illustrate the expected or real situation of the outputs of the postal machine 1 in each of the two sorting cycles making up the sequencing process, as well as the traffic at each delivery point, the load of each postman and the postman who will have to deliver the objects placed at each exit of the postal machine 1.

The balancing method object of the present invention will be described hereinafter with reference to the flow chart illustrated in figures 3a-3k and to a sequencing process having the matrix representation of figure 2.

Furthermore, the balancing method will be described with reference to a sequencing constraint defined by the fact that, at the end of the second sorting cycle, distinct outputs of the postal machine 1 are assigned to the various postmen, so as to avoid the possible confusion that could to be generated in the event that at the end of the sequencing process in an output of the postal machine 1 there should be postal objects relating to different postman paths.

As illustrated in Figure 3a, initially a block 10 is reached in which general information is acquired on the characteristics of the batch to be treated, i.e. the sequencing process to be carried out, such as the number of postmen available for the delivery service, the number of delivery locations to be served, the traffic of postal objects of each delivery location, known for example from historical data, the total number of exits of the postal machine 1, the number of exits of the postal machine 1 to be reserved for emergency situations , for example to be used in place of one or more non-operational exits for the diversion of postal objects, the maximum capacity of each exit of the postal machine 1, a tolerance margin on the capacity of each exit in order to ensure that a desired capacity lower than the maximum capacity of the exit itself, etc.

From block 10 there is then a block 20 in which the number of outlets of the postal machine 1 usable for carrying out the balancing method and the desired capacity for each output of the postal machine 1 is determined.

In particular, the number of outputs that can be used for carrying out the balancing method is determined by subtracting the number of outputs reserved for emergency situations from the total number of outputs of postal machine 1, while the desired capacity for each output of postal machine 1 is determined by subtracting the respective tolerance from the maximum capacity of the output itself.

After the operations described with reference to block 20, a procedure is carried out for verifying the sufficiency of the resources for carrying out the balancing method and the sequencing process.

If this procedure successfully verifies, with the methods described below, the adequacy of the resources for the implementation of the balancing method and the sequencing process, it generates, for example in the memory of the electronic unit 22 that implements it, a matrix representation similar to that of Figure 2 which allows the assignment of the delivery locations of the postal objects to the outlets of the postal machine 1 during the first and second sorting cycle of which the sequencing process is composed.

In particular, block 20 leads to a block 30 in which it is verified whether the number of postmen available for the delivery service is sustainable for the execution of the balancing method, i.e. if the number of postmen available is less than or equal to number of mail machine outputs 1 actually usable for the balancing method.

In particular, if the number of postmen available is less than or equal to the number of outputs that can actually be used for the execution of the balancing method (YES output from block 30) then the method is potentially feasible in relation to the number of postmen who will carry out the distribution and then from block 30 one reaches a block 40, otherwise if the number of available postmen is greater than the number of outputs actually usable for the execution of the balancing method (NO output from block 30) then the balancing method ends for the impossibility of realization due to insufficient number of outlets of postal machine 1 available.

Block 40 determines the minimum number of distinct outlets of the postal machine 1 to be assigned to each postman and the minimum number of outlets that must be used to implement the balancing method.

In particular, the minimum number of distinct outlets to be assigned to each postman is determined on the basis of the number of delivery locations that each postman must serve, the traffic at each delivery location and the number of delivery locations that can be assigned to each machine output. postal 1, equal, as mentioned above, to the maximum number of virtual positions that can be assigned to each output of the postal machine 1 (ie equal to the number of boxes in each column of the matrix of Figure 2).

Once the minimum number of outputs to be assigned to each postman is known, the minimum number of outputs that must be used for the implementation of the balancing method is determined by adding all the minimum numbers of outputs to be assigned to each postman.

From block 40 we then arrive at a block 50 in which it is checked whether the number of outputs of the postal machine 1 that can be used for carrying out the balancing method is sufficient for carrying out the method itself, i.e. whether the number is the number of outputs that can be used for the execution of the balancing method is greater than or equal to the minimum number of outputs that must be used determined in block 40.

In particular, if the number of outputs that can be used for the execution of the balancing method is greater than 0 equal to the minimum number of outputs that must be used (YES output from block 50) then the balancing method is potentially feasible in relation to the number of outputs of the mail machine 1 available and therefore from block 50 you reach a block 60, otherwise if if the number of outputs that can be used for the execution of the balancing method is less than the minimum number of outputs that must be used (NO output from block 50) then the balancing method ends due to the impossibility of realization due to the insufficiency of available outputs.

In block 60 the number of virtual positions available for the realization of the present balancing method is determined by subtracting the number of virtual positions lost (i.e. not usable ) due to the aforementioned constraint on the assignment of separate exits to the various postmen, which can be determined on the basis of the following considerations.

Having assigned to each postman a minimum number of distinct outputs, i.e. a minimum number of distinct columns of the matrix of Figure 2, inevitably each postman has been assigned a number of virtual positions, i.e. a number of boxes in the matrix of Figure 2, equal to an integer multiple of the number of boxes contained in a column, in particular equal to the product between the number of rows in the matrix and the number of distinct columns assigned to it.

Generally, however, the number of virtual positions assigned to each postman is greater than the number of virtual positions that the postman himself would need based on the number of delivery locations assigned to him and consequently a certain number of virtual positions assigned to each postman will never be used by it and will therefore not be used in the balancing method precisely due to the constraint on the assignment of distinct outputs to postmen.

Therefore, the number of virtual positions not used by each postman will be equal to the difference between the number of boxes in the matrix assigned to it and the number of delivery locations it must serve and the total number of virtual positions lost due to the constraint on the assignment. of distinct outputs to postmen will be equal to the sum of the number of virtual positions not used by each of the postmen.

From block 60 we then arrive at a block 70 in which it is verified whether the number of virtual positions available is sufficient for the implementation of the balancing method, i.e. if the number of virtual positions actually available is greater than or equal to the number of delivery locations to serve.

In particular, if the number of virtual positions actually available is greater than or equal to the number of delivery locations to be served (YES exit from block 70) then the balancing method is potentially feasible in relation to the number of virtual positions available and therefore from the block 70 leads to a block 80, otherwise if the number of virtual positions actually available is less than the number of delivery locations to be served (NO exit from block 70) then the balancing method ends due to the impossibility of realization due to insufficient number of virtual positions available.

In block 80 it is checked whether the maximum traffic that can be sequenced by the postal machine 1 is sufficient for carrying out the balancing method, that is, whether the sum of the capacities of the outputs of the postal machine 1 is greater than or equal to the number of postal objects to be sequenced.

In particular, if the maximum sequenceable traffic is greater than the traffic to be sequenced (YES output from block 80) then the balancing method is potentially feasible with respect to the sequencing capacity of the postal machine 1, the resource verification procedure ends and from the block 80, a block 90 is reached, otherwise if the maximum sequenceable traffic is less than the traffic to be sequenced (NO output from block 80) the balancing method ends due to the impossibility of realization due to the insufficiency of addressing capacity of places of address of the postal machine 1.

In block 90 a pre-assignment is made to each postman of a respective output group of the postal machine i on the basis of the number of delivery locations that each of them must serve and the number of virtual positions associated with each output of the postal machine 1.

The matrix of Figure 2 illustrates the situation of the outputs of the postal machine 1 after the pre-assignment step described above; in particular, the matrix only represents the outputs of the postal machine 1 assigned to the postmen while it does not illustrate the outputs yet to be assigned to the postmen and therefore available during the balancing method for assigning delivery locations.

In the event that the balancing method ended during the resource verification procedure, information on the reason why this procedure was abandoned (insufficient number of postmen available, insufficient number of available outlets of postal machine 1, insufficient number of positions virtual, insufficient addressing capacity) procedures are reported so that an employee can act accordingly to modify the initial conditions that did not allow the sequencing process to be carried out.

After the pre-assignment phase, a procedure is carried out for improving the load balancing of each of the outputs of the postal machine 1 at the end of the second sorting cycle of which the sequencing process is composed.

In particular, this procedure operates directly on the matrix of Figure 2 in order to improve the balancing of the load of each of the columns of the matrix itself and provides for the execution of a first and a second sub-procedure which are temporally consecutive.

In detail, block 90 of figure 3a leads to a block 100 of figure 3b in which the first sub-procedure is carried out, which will be described hereinafter with reference to figure 3d.

The first sub-procedure foresees to carry out, for each pair of adjacent columns of each group of columns associated with the postmen, a simulation of movements of particular delivery points between the two columns considered and the actual implementation of the simulated movement which determines the most consistent improvement. load balancing of this pair of columns, with consequent modification of the configuration of the pair of columns considered.

It is evident that if none of the simulated movements improve the load balancing with respect to the initial situation, the configuration of the pair of columns considered remains unchanged.

It should be noted that this movement is carried out in compliance with the initial assignment criterion corresponding to the order of distribution of the postal objects in the delivery points.

In practice, the first sub-procedure moves the delivery locations from the outputs of the postal machine 1 to which they are assigned to respective outputs of the postal machine 1 logically contiguous associated with the same postman, respecting the criterion for assigning the delivery points to the respective groups of outputs.

It is evident that this first sub-procedure is applicable only if the columns associated with a postman are at least two.

In detail, with reference to figure 3d, initially a block 110 is reached in which a pair of adjacent columns to be examined is identified.

In the various iterations, the columns are examined in succession starting from the first two columns of the first group and up to the last two according to their increasing numerical order. It should be noted that the constraint on the assignment of distinct outputs of the postal machine 1 to the various postmen imposes that the adjacent columns identified must belong to the same postman, that is to the same group of columns, as it is only permissible to have an assigned delivery address a postman can be moved to an exit assigned to another postman.

From block 110 one then arrives at a block 120 in which the loads TC1 and TC2 of the columns examined are determined, each of which is equal to the sum of the numerical values represented in the boxes of the respective column. In block 120 a numerical value TM1 is also determined equal to the maximum between the loads TCl and TC2 determined above.

From block 120 one then arrives at a block 130 in which the following two movements are simulated separately:

- the numeric value represented in the occupied box located higher in the first column of the pair (i.e. the left column) is moved to the lower box in the second column of the pair (i.e. the right column) and the numerical values represented in the boxes of this second column before the simulated movement are moved up by one position;

- the numerical value represented in the occupied box located further down in the second column is moved to the free box of the first column placed immediately above the occupied box located higher in the first column itself and the numerical values represented in the other boxes of the second column before the simulated move are shifted down by one position.

These movements are simulated when the boxes of the destination column into which the numerical value is moved are not fully occupied.

In other words, taking into consideration for example the first and second columns of the matrix of figure 2, associated with the postman identified with the letter A, the following two movements are simulated separately:

the numerical value "15" represented in the fourth box of the first column is moved to the first box of the second column and the numerical values "25" and "22" represented in the first and second boxes of the second column before the simulated movement are moved up of a position;

- the numerical value "25" represented in the first box of the second column is moved to the fifth box of the first column and the numerical value "22" represented in the second box of the second column is moved down by one position.

Block 130 then leads to block 140 in which the loads TC3 and TC4 assumed by the two columns examined in the first simulated movement and the loads TC5 and TC6 assumed by the two columns examined in the second simulated movement are determined. Block 140 also determines a numerical value TM2 equal to the maximum between the loads TC3 and TC4 and a numerical value TM3 equal to the maximum between the loads TC5 and TC6.

Block 140 then leads to block 150 in which a numerical value Tm is determined equal to the lower between the numerical value TM1 determined in block 120 and the numerical values TM2 and TM3 relating to the first and second simulated displacements determined in block 140.

If the numerical value Tm is equal to TM2 or TM3 then the simulated displacement of the delivery location from one column to another of the pair that allows to obtain a load equal to Tm on one of the columns examined will correspond to an improvement in the load balancing of the columns examined, while if Tm is equal to TM1 then neither of the two simulated displacements of the delivery location from one column to the other of the pair allows to obtain a load equal to Tm on one of the columns examined will correspond to an improvement in the load balancing of the columns examined and therefore the initial situation before the simulation is the best one.

Therefore, from block 150 one then arrives at block 160 in which it is checked whether Tm is equal to TM1.

If Tm is equal to TM1 (YES output from block 160) then none of the simulated movements are carried out, the configuration of the two columns is not changed and a block 180 is reached, otherwise if Tm is equal to TM2 or TM3 (NO output from block 160) then from block 160 one reaches a block 170 in which the simulated displacement is carried out which allows to obtain the lowest load equal to Tm and the configuration of the two columns is varied according to the simulated displacement -From block 170 it is obtained then to block 180 in which it is checked whether there is a further pair of adjacent columns to be examined in the same group of columns (i.e. associated with the same postman) or, if the columns of a group have already been examined, to the next group of columns (i.e. associated with the next postman).

If there is a further pair of adjacent columns to be examined (YES output from block 180) from block 180 one returns to block 110 for repetition of the first sub-procedure for a subsequent pair of adjacent columns, otherwise if all pairs of adjacent columns are been examined (NO output from block 180) then the first sub-procedure ends.

With reference again to Figure 3b, once the first sub-procedure has been completed, block 100 leads to a block 190 in which it is checked whether the improvement obtained with the first sub-procedure is sufficient or whether a time t0 has elapsed since the first sub-procedure started.

In particular, the improvement obtained is considered sufficient if the loads of each group of columns associated with the various postmen are better balanced, by a predetermined quantity, with respect to the loads that occurred before the execution of the first improvement sub-procedure.

The control of the time elapsed from the beginning of the first improvement sub-procedure, on the other hand, serves to allow the execution of the first sub-procedure only for a predetermined maximum time in order not to excessively occupy the resources that practically implement the present balancing method.

In particular, if the improvement obtained with the first sub-procedure is sufficient or a time tO has elapsed since this sub-procedure started (YES output from block 190) then block 190 leads to a block 200 to carry out assignment operations of further outputs to postmen on the basis of a criticality order of the outputs already assigned to the postmen themselves, otherwise if the improvement obtained with the first sub-procedure is not sufficient or a time tO has not yet elapsed since this sub-procedure began (NO exit from block 190) then from block 190 it goes back to block 100 for the repetition of the first improvement sub-procedure to all the columns of the matrix of figure 2.

In block 200 it is checked whether among the outputs usable for carrying out the balancing method, determined in block 20, there are any available to be assigned.

If among the outputs usable for the execution of the balancing method there are still available to be assigned (output YES from block 200) then from block 200 we reach a block 210, otherwise if among the outputs usable for the execution of the balancing method there are no more than available to assign (NO output from block 200) then a block 240 is reached in which the second aforementioned sub-procedure is carried out, as described in detail below.

In block 210 a criticality order is assigned to the columns which takes into account not only the number of delivery locations associated with each of the columns but also the traffic in each of the delivery locations. In particular, the aforementioned criterion envisages assigning the columns an order of criticality based on the highest number of delivery locations and, with the same delivery locations, on the maximum traffic of the delivery locations.

In other words, the most critical column will be the one having the highest number of occupied cells and the least critical column the one having the lowest number of occupied cells. If two or more columns have the same number of boxes occupied, the most critical column among them will be the one containing the box in which the highest numerical value is represented and the least critical column the one containing the box in which it is represented in value lower number.

From block 210 one reaches a block 220 in which the most critical column is identified on the basis of the assigned criticality order.

From block 220 there is therefore a block 230 in which a free column is assigned to the group of columns containing the most critical column identified above (i.e. to the group of columns assigned to the most critical postman), i.e. a column having all the free boxes.

In practice, in block 230 the most critical postman (ie to whom the most critical column identified above is associated) is assigned a further output of the postal machine 1 to be used for the sequencing process.

The assignment of a further column to the group of columns containing the most critical column is obviously followed by the translation of a position of the columns of the subsequent groups, in the matrix, to the one in which this column has been assigned.

From block 230 one then reaches block 100 again for repetition of the first sub-procedure for improving the balance for all the columns of the matrix of Figure 2, including the one assigned in block 230.

As previously said, if in block 200 it is verified that among the outputs usable for the execution of the balancing method there are no more than available to assign (NO output from block 200) from block 200 one reaches block 240 in which the second sub-procedure performed.

The second sub-procedure involves carrying out, for each pair of adjacent columns of each group of columns associated with the postmen, a simulation of the bisection of particular delivery locations belonging to the columns considered in respective pairs of virtually distinct delivery locations and their differentiated assignment to columns themselves.

According to the second sub-procedure, if a delivery location is bisected, it is removed from the matrix and replaced with two new delivery locations: the first delivery location obtained following bisection is assigned to the same box where the delivery location was. bisected initial and the second place of delivery obtained following bisection is assigned to a box arranged in the column adjacent to the one in which the bisected initial delivery place was located and belonging to the same postman. Furthermore, the traffic of the first and second place of delivery obtained following bisection has a traffic reduced, typically by half, compared to the traffic of the initial delivery place bisected so that their sum is equal to the traffic of the place of delivery initial bisected.

The first and second place of delivery therefore replace the initial bisected place of delivery and actually represent the same postal address, which, through the aforementioned bisection, will be differently and more conveniently assigned to the exits of the postal machine 1 in the first and second sorting cycle. In light of the above, the first and second delivery locations obtained following bisection are therefore virtually distinct.

In particular, each of the bisected delivery locations is arranged, as better described below, in a particular position in the columns considered and is divided into a first and a second virtually distinct delivery location, which are then assigned separately to the columns. same.

Then there is the implementation of the simulated bisection which determines the most consistent improvement of the load balancing of this pair of columns, with consequent modification of the configuration of the pair of columns considered.

It is evident that if none of the simulated bisections determines an improvement in the load balancing, the configuration of the pair of columns considered remains unchanged.

In practice, the bisection of delivery locations into respective pairs of virtually distinct delivery locations is carried out through the second sub-procedure and the assignment of each pair of virtually distinct delivery locations obtained as a result of bisection to a respective pair of outputs of the postal machine 1 logically contiguous, associated with the same postman, comprising the exit to which the delivery location from which said pair of virtually distinct delivery locations originates following bisection is associated and an exit logically contiguous thereto assigned to the same postman.

Furthermore, the assignment of each pair of virtually distinct delivery locations to the respective pair of logically contiguous outlets is carried out by respecting the assignment criterion with which the delivery location from which this pair of virtually distinct delivery locations originates following bisection. is assigned to the respective group of outputs.

It is therefore evident that this second sub-procedure is applicable only if the columns associated with a postman are at least two.

In detail, the second sub-procedure is similar to the first sub-procedure described with reference to the flow chart illustrated in Figure 3d and therefore will be described hereinafter only in so far as it differs from it.

The second sub-procedure differs from the first sub-procedure solely in that the simulations carried out in block 130 are different from those described for the first sub-procedure.

In particular, the following two subdivisions are simulated separately in block 130:

- the numerical value represented in the occupied box located higher up in the first column is divided into a first and a second numerical traffic value which are distinct and quantitatively similar, the sum of which is equal to the original numerical value. The first numerical value is then represented in the box located higher up in the first column in place of the original numerical value while the second numerical value is represented in the box located further down in the second column. The numerical values represented in the boxes of the second column before the simulated movement are then shifted up by one position;

- the numerical value represented in the occupied box located further down in the second column is divided into a first and a second number which are distinct and quantitatively similar, the sum of which is equal to the original numerical value. The first numeric value is then represented in the first free box of the first column arranged immediately above the occupied box arranged higher in the first column itself, while the second numeric value is represented in the box located further down in the second column instead of the value original number.

It is evident that the subdivision of the numerical value represented in the boxes can be carried out in different ways from those described above since it is not absolutely essential for the success of the second sub-procedure that the two numerical values obtained from the subdivision are substantially equal to each other.

During the sorting cycles, the postal object to be delivered to the bisected delivery point will be addressed by the postal machine 1 as if it belonged to one of the two virtually distinct delivery points obtained following bisection, associating it with one or the other other place of delivery with probability proportional to the respective traffic values.

For example, taking into consideration again the first and second columns of the matrix of figure 2, associated with the postman identified with the letter A, the following two simulations are performed separately: the numerical value "15" represented in the fourth box of the first column is divided in a first numerical value equal to "7" and a second numerical value equal to "8". The first numerical value "7" is represented in the fourth box of the first column in place of the original numerical value "15" and the second numerical value "8" is represented in the first box of the second column. The numerical values "25" and "22" represented in the first and second box of the second column before the simulated movement are shifted up by one position.

- the numerical value "25" represented in the first box of the second column is divided into a first numerical value equal to "12" and a second numerical value equal to "13". The first numerical value "12" is represented in the fifth box of the first column and the second numerical value "13" is represented in the first box of the second column in place of the original numerical value "25".

The second sub-procedure allows to obtain, for example, particularly evident results in the improvement of the load balancing when the load of each of the two columns is such that the movements simulated in the first sub-procedure improve the load of the first column but worsen by an approximately equivalent quantity that of the adjacent column and therefore the first sub-procedure does not allow to obtain appreciable improvements.

With reference to Figure 3b, once the second sub-procedure has been completed, block 240 then leads to a block 250 in which it is checked whether the improvement obtained with the second sub-procedure is sufficient or whether a time tl has elapsed since said sub-procedure has started.

If the improvement obtained with the second sub-procedure is significant or a time tl has not elapsed since this sub-procedure has started (NO output from block 250) then from block 250 it goes back to block 240 for repetition of the second sub-procedure, otherwise if the improvement obtained with the second sub-procedure is not significant or a time tl has elapsed since this sub-procedure started (YES output from block 250) then a procedure is performed to improve the balancing of the load of each of the outputs of the postal machine 1 at the end of the first sorting cycle of which the sequencing process is composed.

The procedure for improving the load balancing of each of the outputs of the postal machine 1 at the end of the first sorting cycle operates directly on the matrix of figure 2 in order to improve the load balancing of each of the rows of the matrix itself and provides for the execution of four subprocedures, which are temporally consecutive, described later in the following.

In detail, block 250 of figure 3b leads to a block 300 of figure 3c in which the first improvement sub-procedure is carried out, which will be described hereinafter with reference to figures 3e and 3f.

The first sub-procedure foresees to carry out, for each column of the matrix in figure 2, the simulation of a redistribution of the delivery points of the column on all the boxes of the column itself and the actual implementation of the simulation which determines the most consistent improvement of the balancing of the columns. rows, resulting in a change in the configuration of the column. It is evident that if none of the simulated shifts determines an improvement in load balancing, the configuration of the considered column remains unchanged.

In other words, the first sub-procedure carries out a redistribution of the numerical values represented in the occupied cells of each column on all the cells of the column itself so as to appropriately alternate the free cells and the occupied cells of the columns.

In practice, the first sub-procedure carries out a redistribution of the delivery locations assigned to each output of postal machine 1 at the end of the second sorting cycle on the outputs used in the first sorting cycle, respecting the assignment criteria with which the delivery locations are assigned to the output itself in the second sorting cycle.

It is clear that this first sub-procedure is applicable only if there is at least one free box in the column in which the redistribution is carried out.

In detail, with reference to Figures 3e and 3f, initially we arrive at a block 310 in which the columns of the matrix of Figure 2 are assigned an order of criticality based on a criterion that mainly takes into account the number of delivery locations associated with the columns themselves, i.e. the number of occupied boxes of each column.

Therefore, the most critical column will be the one with the greatest number of occupied cells and the least critical will be the one with the least number of occupied cells.

The order of criticality of the columns is used in the rest of the sub-procedure to distribute the numerical values represented in the boxes of the columns themselves starting from the most critical to the least critical and to distinguish the most critical column from the remaining ones as for it the distribution of the delivery locations on the boxes of the column itself is carried out in a different way than that used for the remaining columns.

Block 310 leads to a block 320 in which the most critical column is identified on the basis of the assigned criticality order.

From block 320 there is then a block 330 in which, for the most critical column, a shift within the column itself of the numerical values represented in the occupied cells is carried out in such a way as to uniformly distribute the free cells among the occupied cells.

For example, taking into consideration the tenth column of the matrix of figure 2, which, as can be seen, is the most critical column, this column has five occupied and five free cells and therefore a uniform distribution between occupied and free cells provides for a shift of the numerical values represented in the occupied cells such as to have a free cell interposed between each pair of occupied cells.

Therefore, the numerical values "26", "48", "13", "11" and "2" represented in the five occupied boxes of the tenth column are thus respectively moved: the numerical value "2" represented in the fifth box is moved to the ninth box, the numeric value "11" represented in the fourth box is moved to the seventh box, the numeric value "13" represented in the third box is moved to the fifth box, the numeric value "48" represented in the second box is moved to the third box and the numerical value "26" represented in the first box remains in that box.

For each of the columns following the most critical one, the distribution takes place according to modalities. different described below.

Therefore, block 330 leads to a block 340 in which a subsequent column to be examined is identified for the distribution of the numerical values represented in the respective occupied boxes starting from the second most critical column and up to the least critical column based on the order of criticality assigned.

Block 340 leads to a block 350 in which the boxes occupied by the column examined are assigned an order of criticality based on a criterion that mainly takes into account the traffic in the delivery points associated with the boxes themselves, i.e. the numerical value represented in the boxes busy.

Therefore, the most critical cell will be the one having the highest numerical value and the least critical will be the one having the lowest numerical value.

Subsequently, operations are carried out with the aim of determining which would be the most convenient shift within the examined column of each of the numerical values represented in the occupied boxes of the column itself for the purpose of minimizing the sum on the rows of the numerical values represented in the occupied boxes belonging to the columns already examined.

In other words, by means of these operations it is determined what would be an optimal position of each of the numerical values within the examined column.

The optimal position of each numerical value represented in the occupied boxes of the examined column is determined by respecting the aforementioned constraint on the progressive order in which the virtual positions are associated with their respective delivery locations (defined, as previously mentioned, by the sequence of adjacent delivery locations corresponding to house numbers, or groups of house numbers, of buildings arranged along the path where the postal objects must be deposited).

In particular, this constraint, prohibiting any exchange of positions between the numerical values represented in the boxes of the same column, requires that the displacement of the numeric value represented in a box corresponds, in case of overlap with another numeric value, an equivalent displacement of the values numbers represented in the contiguous boxes of the same column placed above the box (if moving upwards) or below the box (if moving downwards).

Therefore, to determine the optimal position of each of the numerical values represented in the occupied boxes of the examined column, block 350 leads to a block 360 in which, in the examined column, an occupied box is identified to be examined starting from the most critical box. and, in subsequent repetitions, up to the least critical box based on the previously assigned criticality order.

From block 360 one then reaches a block 370 in which the load TC of the row RC in which the examined box is arranged is determined, considering, as previously said, only the numerical values of the row represented in the boxes of the columns already examined.

From block 370 one then arrives at a block 380 in which the possible displacements to be implemented of the numerical value represented in the box examined in the respective column are determined, i.e. the displacements for which it is possible to carry out a corresponding displacement of the numerical values represented in the other boxes occupied by the column itself.

From the biotic 380 we then arrive at a block 390 in which it is verified whether there is an actionable shift to be simulated within the same column of the numerical value represented in the examined box.

If there is an actuable movement to simulate (YES output from block 390) then from block 390 one arrives at a block 400, otherwise if there is no actuable movement to simulate (NO output from block 390) then from block 390 one arrives at a block 470 described below.

In block 400 the sums per row of the numerical values represented in the boxes belonging to the columns already examined are determined, excluding the column currently examined and limited to the rows on which it is possible to move the numerical value represented in the examined box and also including the row a to which the checked box belongs.

From block 400 one then arrives at a block 410 in which, among the lines taken into consideration in block 400, the line having the minimum sum of traffic is determined.

From block 410 one then reaches a block 460 in which the above-identified shift is carried out, consequently carrying out, when necessary, the corresponding displacements of the other numerical values represented in the other occupied cells of the column.

It is evident that no movement will be made if the row to which the numerical value represented in the examined box initially belongs is the one with the minimum sum of traffic.

From block 460 we then arrive at a block 470 in which the numerical value represented in the examined box is constrained to assume the assumed position, be it the initial one or the one in which it was moved, so that it is no longer subject to further movements in the subsequent simulations carried out for the other numerical values represented in the other boxes of the column examined.

Taking, for example, into consideration the ninth column of the matrix of Figure 2, which as can be seen is the second most critical, the most critical box is the first in which the numerical value "64" is represented. As described above, this numerical value is shifted to one of the second, fourth, sixth and eighth rows of the column itself.

Winning over the progressive order of the virtual positions does not allow, for example, the displacement of the numerical value "64" in the tenth box since the numerical value "30" represented in the second box cannot be moved accordingly in the same column.

From block 470 one then reaches block 480 in which it is checked whether there is a further occupied box to be examined in the examined column on which to simulate the movement and repeat the operations described for the numerical value represented in that box.

If there is a further occupied box to be examined (YES output from block 480) then from block 480 one reaches block 360 again to repeat the operations described for the numerical value represented in the next occupied box examined with the constraint of not moving the numerical values represented in the boxes already examined, otherwise if all the occupied boxes of the column examined have already been examined (NO output from block 480) then block 480 leads to block 490.

In block 490 it is checked whether there is a further column to be examined in the matrix of figure 2.

If there is a further column to be examined (YES output from block 490) then from block 490 one returns to block 340 to repeat the operations described for the occupied cells of the next column examined, otherwise if all the columns of the matrix have already been examined (NO output from block 490) then the first sub-procedure ends.

With reference again to Figure 3c, once the first sub-procedure has been completed, block 300 leads to block 530 in which the second sub-procedure is carried out, which will be described hereinafter with reference to Figures 3g and 3h.

The second sub-procedure provides for carrying out, for each row of the matrix of Figure 2 and according to an order described in detail below, the simulation of the displacement of each of the numerical values of the row itself in free contiguous boxes of the same column and its actual implementation only if this move allows to obtain improvements on the balancing of the rows.

In other words, the second sub-procedure simulates the shifting of each of the numerical values represented in the occupied boxes of each row into adjacent boxes of the same column.

In practice, the second sub-procedure moves delivery locations from the exits of postal machine 1 to which they are assigned to respective logically contiguous exits at the end of the first sorting cycle, respecting the assignment criterion with which the delivery locations are assigned to respective groups of outlets in the second sorting cycle.

It is evident that this second sub-procedure is applicable only if at least one of the occupied cells has at least one adjacent free cell.

In detail, with reference to Figures 3g and 3h, initially we arrive at a block 550 in which the rows of the matrix of Figure 2 are assigned an order of criticality based on a criterion that mainly takes into account the number of delivery locations associated with rows themselves, i.e. the number of occupied boxes of each row.

Therefore, the most critical line will be the one with the greatest number of occupied cells and the least critical will be the one with the least number of occupied cells.

The order of criticality of the rows is used to determine the most critical row on which to simulate the displacement of the numerical values represented in the boxes of the row itself.

Block 550 leads to a block 560 in which the most critical row RC to be examined for the displacement of the numerical values represented in its own occupied cells is identified.

From block 560 one then arrives at a block 570 in which the load TC of the row RC examined is determined by adding the numerical values represented in the occupied boxes of the row itself.

Subsequently, operations are carried out with the aim of determining which would be, for the purpose of minimizing the load on the rows, the most convenient movement within the same column of each of the numerical values represented in the occupied boxes of the row examined RC from the position occupied in the row RC itself to the positions in the RT lines immediately contiguous to the RC line.

In other words, by means of these operations it is determined what would be the optimal position, in order to minimize the load of the rows, of each of the numerical values represented in the occupied boxes of the RC row examined between the original position in the RC row itself, the one immediately beyond above and the one immediately below in the same column.

It is evident that for a given numerical value these operations can only be carried out if in at least one of the rows immediately contiguous to the one in which this numerical value is represented there is a free box in the same column.

Therefore, to determine the optimal position of each of the numerical values represented in the occupied boxes of the row RC examined, block 570 leads to a block 580 in which, in the row RC examined, an occupied box to be examined is identified, starting from box arranged in the first column and up to the one arranged in the last column.

From block 580 one then arrives at a block 590 in which the possible displacements to be implemented are determined that the numerical value represented in the examined box can carry out within the same column from row RC to contiguous rows RT, i.e. the displacements that this numerical value he can carry out to free squares contiguous to the one examined.

It is evident that these movements can be at most two as there are two boxes adjacent to the one considered.

From block 590 one then arrives at a block 600 in which it is checked whether there is an actionable shift to simulate of this numerical value from row RC to a contiguous row RT.

If there is an actuable movement to simulate (output YES from block 600) then block 600 leads to a block 610, otherwise if all the actuable movements have already been simulated (NO output from block 600) then block 600 leads to a block 660 described below.

In block 610 one of the displacements determined in block 590 of the numerical value represented in the box examined by the row RC is simulated to one of the contiguous rows RT and the new load TCN of the row RC and the new load TTN of the row RT after the displacement are determined.

From block 610 one then arrives at a block 620 in which a numerical value TM equal to the maximum between the two loads TCN and TTN determined above is determined, ie TM = max (TCN, TTN).

From block 620 one then arrives at a block 630 in which it is checked whether the numerical value TM is less than the load TC of the row RC examined before the simulated movement and determined in block 570, i.e. it is checked whether the simulated movement has determined improvements in the balance of the load of the RC and RT lines.

If the numerical value TM is less than the load TC, that is, if the simulated movement has determined improvements in the load balancing of the rows RC and RT (output YES from block 630) then from block 630 a block 640 is reached, otherwise if the value numerical TM is greater than or equal to the load TC, i.e. if the simulated movement did not determine improvements in the load balancing of the rows RC and RT (output NO from block 630) then from block 630 one returns to block 600 again to check whether there is a further displacement which can be simulated from row RC to a contiguous RT and repeat the operations described above in relation to said displacement.

In block 640 the difference D between the load TC and the numerical value TM is determined, which is correlated to the extent of the improvement in the load of the rows RC and RT obtained by means of the displacement.

From block 640 one then reaches a block 650 in which the above determined distance D and the relative displacement which allows to obtain this difference D, as well as the position of the box examined, are stored.

From block 650 one then returns to block 600 to check if there is a further feasible shift to be simulated from row RC to a contiguous RT and to repeat the operations described above in relation to this shift.

In block 660, in which it is reached whether there are no feasible displacements to simulate or if all the possible displacements to be implemented of the numerical value represented in the examined box have been simulated, it is checked whether there is a further box to be examined belonging to the RC row examined .

If there is a further box to be examined (YES exit from block 660) then from block 660 one returns to block 580 for the repetition of the operations described for a subsequent occupied box of the RC row, otherwise if all the boxes have been examined (exit NO from block 660) then a block 670 is reached.

In block 670 it is checked whether at least one displacement and a respective distance D relative to the examined row RC have been stored, that is, if at least one of the simulated displacements has made it possible to obtain an improvement in the load balancing of the rows RC and RT.

If at least one movement has been memorized, i.e. at least one of the simulated movements has allowed to minimize the load of the rows RC and RT (output YES from block 670), from block 670 one reaches a block 680 otherwise if no movement has been memorized, that is, no simulated movement has allowed to minimize the load of the rows (NO output from block 670), from block 670 one reaches a block 700.

Block 680 identifies which of the stored distances D is the maximum and consequently also identifies the displacement which allows to obtain this maximum distance D as well as the position of the box affected by this displacement.

From block 680 one then arrives at a block 690 in which the above-identified displacement is carried out, with consequent modification of the configuration of the matrix of Figure 2.

From block 690, block 550 is then reached again for a new assignment of an order of criticality to the rows and the repetition of the operations described above for the most critical row.

In block 700, in which it is reached if no movement has been stored, that is, if no movement in the row examined has allowed an improvement in the load balancing of the rows, the row RC previously examined is considered excluded from the operations carried out subsequently on the rows of the matrix of figure 2 since any shift of the numerical values of its boxes does not allow to obtain any improvement on the balancing of the rows themselves.

From block 700 one then arrives at a block 710 in which it is checked whether all the lines have been excluded or whether a predetermined maximum time has elapsed. If not all the rows have been excluded and a predetermined maximum time has not yet elapsed (YES output from block 710), from block 710 it goes back to block 550 for a new assignment of an order of criticality to the rows and the repetition of the operations described above in the most critical line, otherwise if all the lines have been excluded or not all the lines have been excluded but a predetermined maximum time has elapsed - (NO output from block 710) then the second sub-procedure ends.

With reference again to Figure 3c, once the second sub-procedure has been completed, block 530 leads to a block 720 in which it is checked whether the improvement obtained with the second sub-procedure is sufficient or whether a time t2 has elapsed since said second sub-procedure has started.

In particular, the improvement obtained is sufficient if the loads of each group of rows associated with the various postmen are better balanced, by a predetermined quantity, with respect to the loads that occurred before the execution of the second improvement sub-procedure.

If the improvement obtained with the second improvement sub-procedure is not sufficient or a time t2 has not elapsed since this second sub-procedure started (NO output from block 720) then block 720 returns to block 530 for repetition of the second sub-procedure , otherwise if the improvement obtained with the second improvement sub-procedure is sufficient or a time t2 has elapsed since the second sub-procedure started (YES output from block 720) then block 720 leads to a block 730.

The third sub-procedure is carried out in block 730, which will be described hereinafter with reference to Figures 3i, 3j and 3k.

The third sub-procedure envisages carrying out a simulation of exchanges between contiguous rows of pairs of delivery locations belonging to different columns and carrying out such exchanges only if it allows to obtain improvements on the load balancing of said pair of columns.

In other words, for each of the numerical values represented in the RC row examined, the third sub-procedure simulates the shift, within a first column, of a numerical value represented in an occupied box of the RC row into a free box of a RT row. immediately contiguous and the simultaneous displacement, within a second column, of a numerical value represented in an occupied box of row RT in a free box of row RC.

In practice, the third sub-procedure carries out exchanges of delivery points between contiguous outlets and each exchange is carried out respecting the assignment criterion with which the exchanged delivery points are assigned to the respective groups of outlets in the second sorting cycle.

It is clear that this third sub-procedure is applicable only if there is at least one free box within each of the aforementioned columns.

In detail, with reference to figures 3i, 3j and 3k, initially we arrive at a block 750 in which the rows of the matrix of figure 2 are assigned an order of criticality based on a criterion that mainly takes into account the number of delivery locations associated with the rows themselves, i.e. the number of occupied boxes of each row.

Therefore, the most critical line will be the one with the greatest number of occupied cells and the least critical will be the one with the least number of occupied cells.

The order of .criticity of the rows is used to determine the most critical row to be considered for the simulation of the exchange of pairs of numerical values. From block 750 it reaches a block 760 in which the most critical row RC to be examined for the exchange of the numerical values represented in its own occupied cells is identified.

From block 760 one then arrives at a block 770 in which the load TC of the row RC examined is determined by adding the numerical values represented in the occupied boxes of the row itself.

Subsequently, operations are carried out with the aim of determining which would be, for the purpose of minimizing the load on the rows, the most convenient movement within the same column of each of the numerical values represented in the occupied boxes of the RC row examined by the position in the RC row to a position in the immediately contiguous row RT and a simultaneous shift in the opposite direction of another numerical value from the position in the RT row to the position in the RC row, obviously placed in another column.

In other words, by means of these operations it is determined which would be the most convenient exchange of position, in order to minimize the load of the rows, of each of the numerical values represented in the occupied boxes of the row examined with another numerical value placed in a row immediately contiguous.

It is evident that for a given numerical value these operations can only be carried out if in at least one of the lines immediately above or immediately below the box in which this numerical value is represented there are free boxes.

Therefore, to determine the most convenient exchange of each of the numerical values represented in the occupied boxes of the row RC examined, block 770 leads to a block 780 in which, in the row RC examined, an occupied box to be examined is identified, starting from the box in the first column to the box in the last column.

From block 780 one then arrives at a block 790 in which the possible displacements to be implemented are determined that the numerical value represented in the examined box can carry out within the respective column on rows RT contiguous to row RC, i.e. the displacements towards free contiguous boxes .

It is evident that these movements can be at most two as there are two boxes adjacent to the one considered.

From block 790 one then arrives at a block 800 in which it is checked whether there is an actionable shift to simulate of this numerical value from row RC to a contiguous row RT.

If there is an actuable movement to simulate (YES output from block 800) then block 800 leads to a block 810, otherwise if there is no actionable movement to simulate or all the movements have already been simulated (NO output from block 800 ) then block 800 leads to a block 900 described below.

In block 810 one of the displacements determined in block 790 of the numerical value represented in the box examined by the row RC is simulated to one of the contiguous rows RT and the load TT of the row RT is also determined before the displacement.

From block 810 one then arrives at a block 820 in which it is determined which of the numerical values represented in the occupied cells of row RT can be moved, by moving them within their column, into free cells of row RC.

From block 820 one then arrives at a block 830 in which it is checked whether there is an actionable shift to simulate of the numerical values represented in the occupied boxes of the RT line from the RT line itself to the RC line.

If there is an actuable movement to simulate (YES output from block 830) then block 830 leads to a block 840, otherwise if there is no actionable movement to simulate or all the movements have already been simulated (NO output from block 830 ) then from block 830 you go back to block 800 to check if there is a further feasible shift to simulate of the numerical value represented in the occupied box examined by the row RC to another of the rows RT contiguous to the row RC itself and then repeat the operations described above in relation to this further displacement.

In block 840 the movement of one of the numerical values, determined in block 820, from line RT to line RC is simulated.

From block 840 one then reaches a block 850 in which the new load TCN of the row RC after the simulated movement and the new load TTN of the row RT after the simulated movement are determined.

From block 850 one then arrives at a block 860 in which a first numerical value TM1 is determined equal to the maximum between the two loads TC and TT of the lines RC and RT before the movements simulated and determined in blocks 710 and 810, i.e. TMl = max (TC, TT), and a second numerical value TM2 equal to the maximum between the two loads TCN and TTN of the lines RC and RT after the movements simulated and determined in block 840.

From block 860 we then arrive at a block 870 in which it is verified whether the second numerical value TM2 is less than the first numerical value TM1, i.e. if the exchange of numerical values between the rows RC and RT simulated above has determined an improvement in the loads of the RC and RT lines themselves.

If the second numerical value TM2 is less than the first numerical value TM1, that is, if the exchange has determined an improvement in the balancing of the loads of the RC and RT lines (output YES from block 870) then from block 870 a block 880 is reached, otherwise if the second numerical value TM2 is greater than or equal to the first numerical value TM1, that is, if the exchange has not determined an improvement in the balancing of the loads of the RC and RT lines (NO output from block 870) then block 870 returns to the block 830 to verify if there is a further feasible shift of a numerical value from row RT to row RC to be simulated and repeat the operations described above for such further displacement.

In block 880 the difference D between the first and the second numerical value TM1 and TM2 is determined, which is correlated to the extent of the improvement in the load of the rows RC and RT obtained by means of the exchange.

From block 880 one then arrives at a block 890 in which the above-determined distance D and the relative pair of displacements that allow to obtain this difference D, as well as the positions of the boxes affected by the aforementioned displacements, are stored.

From block 890 one then reaches block 830 again to check if there is a further feasible shift of a numerical value to be simulated from line RT to line RC and repeat the operations described above for this further shift.

As previously said, if there is a further feasible shift to simulate from row RT to row RC (YES output from block 830) then block 830 leads to block 840, otherwise if there is no feasible displacement to simulate or all displacements that can be implemented from line RT to line RC have been simulated (NO output from block 830) from block 830 you reach block 800 to check if there is a further feasible shift to be simulated from line RC to line RT and repeat the operations described above in relation to this further shift.

If there is a further feasible shift to be simulated from row RC to row RT (YES output from block 800) then, as previously said, from block 800 one reaches block 810, otherwise if there is no feasible shift to simulate or all the movements that can be carried out from row RC to row RT have been simulated (NO output from block 800) from block 800 one reaches block 900.

In block 900 it is checked whether there is a further box of the RC row to be examined.

If there is a further box of the RC line to be examined (YES output from block 900) then from block 900 it goes back to block 780 for the repetition of the operations described for a subsequent occupied box of the RC line, otherwise if all the boxes of the row RC have been examined (NO output from block 900) then a block 910 is reached.

In block 910 it is checked whether at least one pair of displacements and a respective distance D relative to the examined row RC have been stored, that is, if at least one of the simulated exchanges has made it possible to obtain an improvement in the load balancing of the rows RC and RT.

If at least one pair of movements has been stored, i.e. at least one of the simulated exchanges has allowed to minimize the load of the RC and RT lines (output YES from block 910), block 910 leads to a block 920, otherwise if no pair of displacements has been memorized, ie none of the simulated exchanges allowed to minimize the load of the rows RC and RT (NO output from block 910), from block 910 one reaches a block 940.

Block 920 identifies which of the memorized distances D is the maximum and consequently also identifies the pair of displacements which makes it possible to obtain such distance D as well as the positions of the boxes affected by such displacements.

From block 920 one then arrives at a block 930 in which the pair of displacements identified above is implemented, with consequent modification of the configuration of the matrix of Figure 2.

From block 930, block 750 is then reached again for a new assignment of an order of criticality to the rows and the repetition of the operations described above for the most critical row.

In block 940, in which it is reached if no pair of displacements have been stored, i.e. if no simulated displacement allows to obtain improvements in the load balancing of the RC and RT rows, the row RC previously examined is considered excluded from the operations carried out subsequently on the rows of the matrix of figure 2 since any shift of the numerical values of its boxes does not allow to obtain any improvement on the balancing of the rows themselves.

From block 940 one then reaches a block 950 in which it is checked whether all the rows have been excluded or whether a predetermined maximum time has elapsed.

If not all the rows have been excluded and a predetermined maximum time has not yet elapsed (YES output from block 950), from block 950 it goes back to block 750 for a new assignment of an order of criticality to the rows and the repetition of the operations described above for the most critical line, otherwise if all the lines have been excluded or not all the lines have been excluded but a predetermined maximum time has elapsed (NO output from block 950) then the third sub-procedure ends.

With reference again to Figure 3c, once the third sub-procedure has been completed, block 730 leads to a block 970 in which it is checked whether the improvement obtained with the third sub-procedure is sufficient or whether a time t3 has elapsed since said third sub-procedure has started.

In particular, the improvement obtained is sufficient if the loads of each group of rows associated with the various postmen are better balanced, by a predetermined quantity, with respect to the loads that occurred before the execution of the third improvement sub-procedure.

If the improvement obtained with the third improvement sub-procedure is significant, i.e. higher than a predefined threshold, or if a time t3 has not elapsed since this third sub-procedure started (NO output from block 970) then block 970 returns to block 730 for the repetition of the third sub-procedure, otherwise if the improvement obtained with the third improvement sub-procedure is not significant or a time t3 has elapsed since this third sub-procedure has started (YES output from block 970) then block 970 leads to a block 980.

The fourth sub-procedure is carried out in block 980, which provides for carrying out, for each of the rows of the matrix of Figure 2 according to a predetermined order described in detail below, a simulation of the subdivision (bisection) of each of the delivery locations arranged in such rows in two virtually distinct delivery locations and their separate assignment to nearby free boxes of the same column and an implementation of this subdivision only if it allows to obtain improvements on the load balancing of said pair of rows.

According to the fourth sub-procedure, therefore, if a delivery location is bisected, it is removed from the matrix and replaced with two new delivery locations: the first delivery location obtained following bisection is assigned to the same box where the location was bisected initial delivery address and the second delivery address obtained following bisection is assigned to a box arranged in the same column in which the bisected initial delivery address was arranged but in a neighboring row, compatibly with the free boxes available in the matrix. Furthermore, the traffic of the first and second delivery points obtained following bisection has a traffic reduced, typically by half, compared to the traffic of the initial bisected delivery place, so that their sum is equal to the traffic of the place of delivery. initial address bisected.

The first and second place of delivery therefore replace the initial bisected place of delivery and actually represent the same postal address, which, through the aforementioned bisection, will be differently and more conveniently assigned to the exits of the postal machine 1 in the first and second sorting cycle. In light of the above, the first and second delivery locations obtained following bisection are therefore virtually distinct.

Through the fourth sub-procedure, a bisection of delivery locations is carried out in respective pairs of virtually distinct delivery locations and each pair of virtually distinct delivery locations obtained as a result of bisection is assigned to a respective pair of outputs including the output to which it is associated with the place of delivery, from which this pair of virtually distinct delivery places originates following bisection, and a second exit identified by the position to which the second virtual delivery place is assigned.

Furthermore, the assignment of each pair of virtually distinct delivery locations to the respective pair of exits is carried out respecting the assignment of the delivery location from which this pair of virtually distinct delivery locations originates following bisection at the respective exit in the second. sorting cycle.

It is clear that this fourth sub-procedure is applicable only if at least one of the boxes next to the one examined is free.

The fourth sub-procedure is similar to the second sub-procedure described above with reference to Figures 3g and 3h and therefore will be described only in what differs from the second sub-procedure.

The fourth sub-procedure differs from the second sub-procedure only in that instead of simulating displacements on contiguous lines of the numerical value represented in the occupied box examined, the subdivision of this numerical value into a first and a second numerical value distinct and quantitatively similar is simulated, whose sum is equal to the original numerical value.

The first numeric value is then represented in the box in which the original numeric value is placed, while the second numeric value is represented in a nearby free box of the same column.

It is evident that the subdivision of the original numerical value can be carried out in different ways from the one described above since it is not at all essential for the success of the second sub-procedure that the two numerical values obtained from the subdivision are quantitatively similar to each other.

At the end of the simulation of the bisection of all the numerical values represented in the occupied boxes of the examined row, the subdivision to which the maximum difference D is associated is then implemented.

At the end of the fourth sub-procedure, block 980 of figure 3c leads to a block 990 in which it is checked whether the improvement obtained with the fourth sub-procedure is significant or whether a time t4 has elapsed since said fourth sub-procedure has started.

In particular, the improvement obtained is considered significant if the loads of each group of rows associated with the various postmen are better balanced, by a quantity greater than a predetermined threshold, with respect to the loads that occurred before the execution of the third improvement sub-procedure.

If the improvement obtained with the fourth sub-procedure is significant or a time t4 has not elapsed since this fourth sub-procedure started (NO output from block 990) then from block 990 it goes back to block 980 for the repetition of the fourth sub-procedure, otherwise if the improvement obtained with the fourth sub-procedure is not significant or a time t4 has elapsed since said third sub-procedure has started (YES output from block 990) then the balancing method ends.

As initially said, the balancing method described above relates to a postal machine 1 having only one input I.

In the event that the postal machine 1 has several inputs for the postal objects, the sequencing process is similar to that described and differs from this in the following.

In particular, if the postal machine 1 had several entrances, each of these would be assigned a first subset of the exits at the end of the first sorting cycle and a second subset of the exits at the end of the second sorting cycle.

In other words, the postal objects entering the postal machine 1 through the first entrance would be deposited in a first subset of the outputs of the postal machine 1 itself at the end of the first sorting cycle and in a second subset of the outputs of the postal machine 1 at the end of the second sorting cycle.

A similar argument applies to postal objects entering the postal machine 1 through the second entrance, the third entrance, and so on.

Therefore, by identifying the first and second subset of the outputs of the postal machine 1 associated with each input on the rows and, respectively, on the columns of the matrix of Figure 2, groups of rows and respective groups of columns associated with each input of the postal machine 1 at the end of the first and, respectively, of the second sorting cycle.

By determining the intersection of each group of rows with the respective group of columns, distinct groups of boxes of the matrix of Figure 2 are identified, each of which defines a respective submatrix contained in the matrix of Figure 2.

Each submatrix therefore constitutes the matrix representation of a postal machine in all respects having an input and the related subset of outputs and can therefore be treated as the matrix of figure 2 without changes with respect to what was previously described, thus allowing to improve the balancing of the subset of outputs associated with the respective input.

As initially said, the balancing method has been described with reference to a postal machine 1 carrying out a sequencing process consisting of two successive sorting cycles.

However, the inventive concept underlying the present invention can also be used in a sequencing process consisting of several successive sorting cycles.

In this case, in the last sorting cycle the first procedure will be applied as described, in the first sorting cycle the second procedure will be applied as described, while in the second sorting cycle (intermediate sorting cycle) it can be chosen not to apply none of the above procedures or apply one or both of them.

As initially said, the method according to the present invention can also be used for load balancing in any sequencing process carried out by means of one or more devices capable of separating streams of objects to be arranged in a predetermined sequence for addressing in places address or destination.

Therefore, the term "machine" used in the previous description generically indicates, in these different uses of the present balancing method, the device or devices used for carrying out the sequencing process.

In particular, in these different uses of the present balancing method, the machine is able to pick up in an orderly way from an input, consisting of a generic number of waiting areas, a sequence of objects whose order can be unknown a priori and each of the which contains sufficient elements for a classification by the machine itself and to guarantee the transport of the objects themselves to an exit consisting of storage and / or release areas, which are dependent on the identified class of the object according to a pre-established rule and they can also coincide with the aforementioned waiting areas.

From an examination of the characteristics of the balancing method according to the present invention, the advantages that it allows to be obtained are evident.

Finally, it is clear that modifications and variations can be made to the method described and illustrated here without thereby departing from the protective scope of the present invention.

For example, the criticality criteria adopted for sorting the rows and columns and the criteria on the basis of which the various sub-procedures end may be different from those described.

Also, in the sub-procedures where no critical ordering criteria are applied, the method by which the rows and columns are examined may be different from those described.

Furthermore, in the sub-procedures described with reference to blocks 530, 730 and 980, the displacements and exchanges between numerical values can be carried out in different ways from those described, and in particular, if possible, they can also be carried out in positions that are not immediately contiguous but two distant. or more boxes.

The same argument can be made for the sub-procedure described in block 980 and therefore, if it is possible, the virtually distinct numerical values obtained following bisection can be assigned to boxes placed in rows that are not immediately contiguous.

Furthermore, the criteria for assigning delivery locations to the boxes in the matrix may be different from that described.

The use of an assignment criterion different from the one described results in a series of variations on the methods of moving the places of delivery.

In particular, as previously mentioned, the criterion for assigning delivery locations to the boxes of the matrix defines a constraint prohibiting any exchange of positions between the numerical values represented in the boxes of a column and this constraint requires that the various displacements and bisections of numerical values described above in the sub-procedures illustrated in blocks 100 and 240 are carried out in compliance with this constraint.

Therefore, by using an assignment criterion different from the one described, it is evident that any displacement of numerical values must comply with this different assignment criterion and therefore the displacements will be made between columns considered "contiguous" according to the allocation criterion adopted.

Consequently, also the "logical contiguity" of the mail machine outputs, which in the example described is defined by a memorized table and is unique in all sorting cycles, and also at the level of the various sub-procedures, could be different and in each sub-procedure of the balancing method the outputs of the mail machine could be contiguous to each other according to a different logical criterion.

Finally, the load balancing of the postal machine outputs could be simplified by performing even one of the aforementioned first and second balancing procedures in the respective sorting cycle and this simplification, even if on the one hand it would determine an inevitable reduction in the effectiveness of the balancing of the load of the outputs of the postal machine, however, would allow on the other hand a considerable saving in execution time and in the occupation of the resources required for the implementation of the method itself.

Claims (1)

R I V E N D I C A Z I O N I 1. - Method of load balancing in a process of sequencing objects (2), in particular postal objects, said sequencing process comprising the execution of a first and at least a second consecutive sorting cycle and being carried out by means of a machine ( 1) receiving in input (I) a set of objects (2) and supplying on its outputs (U) said objects (2) identified and separated according to determined sorting rules; in a given sorting cycle the objects (2) being supplied to the outlets (U) of the machine (1) according to a respective predetermined sorting criterion and being neatly supplied again at the input (I) to the machine (1) itself for execution of a subsequent sorting cycle; the outputs (U) of the machine (1) being divided, in the last sorting cycle of the Sequencing process, into distinct subsets (A, B, C, D) and to each subset (A, B, C, D) being assigned a plurality of respective places of destination according to an assignment criterion correlated to the order of distribution of the objects in the places of destination; said balancing method being characterized in that it comprises the step of carrying out, in at least one of said sorting cycles, at least one of the following procedures: a) a first procedure (100, 210-240) for balancing the load of the outputs (U) of the machine (1) at the end of the current sorting cycle based on at least one of the following balancing criteria: • displacement of destination locations from the exits (U) to which they are assigned to respective logically contiguous exits (U) according to a first predetermined logical criterion and belonging to the same subset (A, B, C, D); • assignment of further outputs (U) to said subsets (A, B, C, D) based on the criticality order of the outputs already assigned to the subsets (A, B, C, D) themselves; And • bisection of places of destination into respective pairs of virtually distinct places of destination and assignment of each pair of virtually distinct places of destination to a respective pair of logically contiguous outputs (U) according to a second predetermined logical criterion and belonging to the same subset ( A, B, C, D); and b) a second procedure (300, 530, 730, 980) for balancing the load of the outputs (U) of the machine (1) at the end of the current sorting cycle based on at least one of the following balancing criteria: • distribution on the outlets (U) used in the current sorting cycle of places of destination assigned to each outlet (U) at the end of the next sorting cycle; • displacement of destinations from the exits (U) to which they are assigned to respective logically contiguous exits (U) according to a third pre-established logical criterion; <■> exchange of destination places between logically contiguous outputs (U) according to a fourth predetermined logical criterion; And • bisection of destination places into respective pairs of virtually distinct destination places and assignment of each pair of virtually distinct destination places to a respective pair of logically contiguous outputs (U) according to a fifth predetermined logical criterion. 2.- Method according to Claim 1, characterized by the fact that said first procedure (100, 210-240) comprises the step of carrying out a first sub-procedure (100) for moving destination locations from the outputs to which they are assigned to respective logically contiguous outputs belonging to the same subset (A, B, C, D); said first sub-procedure (100) comprising the step of moving the places of destination respecting the assignment criterion with which the places of destination themselves are assigned to the respective subsets (A, B, C, D) of outputs (U) in the last cycle sorting of the sequencing process. 3.- Method according to Claim 1 or 2, characterized in that said first procedure (100, 210-240) comprises the step of carrying out a second sub-procedure (210-230) for assigning further outputs (U) to said subsets ( A, B, C, D) on the basis of a criticality order of the outputs (U) already assigned to the subsets (A, B, C, D) themselves; said order of criticality being a function of the number of places of destination assigned to each exit (U) and of the traffic of objects in each place of destination. 4.- Method according to any one of the preceding claims, characterized in that said first procedure (100, 210-240) comprises the step of carrying out a third sub-procedure (240) of bisection of places of destination in respective pairs of places of destination virtually distinct and assignment of each pair of virtually distinct destination locations to a respective pair of logically contiguous outputs (U) belonging to the same subset (A, Β, C, D); said third sub-procedure (240) comprising the step of assigning each pair of virtually distinct destination places to a logically contiguous pair of exits (U) comprising the exit to which the destination place is associated from which said pair of destination places virtually distinct originates as a result of bisection and an exit (U) logically contiguous to it belonging to the same subset (A, B, C, D). 5.- Method according to Claim 4, characterized in that said third sub-procedure (240) comprises the step of assigning each pair of virtually distinct destination sites obtained following bisection to the respective pair of logically contiguous outputs (U) respecting the criterion assignment with which the place of destination from which this pair of virtually distinct places of destination originates following bisection is assigned to the respective subset (A, B, C, D) of outputs (U) in the last sorting cycle of the sequencing process. 6.- Method according to any one of the preceding claims, characterized in that said second procedure (300, 530, 730, 980) comprises the step of carrying out a fourth distribution sub-procedure (300) on the outlets (U) used in the sorting cycle current of destination locations assigned to each exit (U) at the end of the next sorting cycle; said fourth sub-procedure (300) comprising the step of distributing the places of destination respecting the assignment criterion with which the places of destination themselves are assigned to the relative output in the last sorting cycle of the sequencing process. 7.- Method according to any one of the preceding claims, characterized in that said second procedure (300, 530, 730, 980) comprises the step of carrying out a fifth sub-procedure (530) for moving destination locations from the exits to which they are assigned to respective logically contiguous outputs; said fifth sub-procedure (530) comprising the step of moving the places of destination respecting the assignment criterion with which the places of destination themselves are assigned to the respective subsets of outputs in the last sorting cycle of the sequencing process. 8. A method according to any one of the preceding claims, characterized in that said second procedure (300, 530, 730, 980) comprises the step of carrying out a sixth sub-procedure (730) for exchanging places of destination between logically contiguous exits; said sixth sub-procedure (730) comprising the step of carrying out each exchange by moving a first destination location from a first output to which it is assigned to a logically contiguous second output and simultaneously moving a second destination location from said second output to said first output; each of said exchanges being carried out respecting the assignment criteria with which the first and second place of destination are assigned to the respective subsets of outputs in the last sorting cycle of the sequencing process. 9.- Method according to any one of the preceding claims, characterized in that said second procedure (300, 530, 730, 980) comprises the step of carrying out a seventh sub-procedure (980) for bisecting destination sites in respective pairs of places of destination virtually distinct destination and assignment of each pair of virtually distinct destination places obtained following bisection to a respective pair of logically contiguous outputs; said seventh sub-procedure (980) comprising the step of assigning each pair of virtually distinct destination places to a logically contiguous pair of exits comprising the exit to which the destination place is associated from which said pair of virtually distinct destination places originates following bisection and an exit logically contiguous to it. 10.- Method according to Claim 9, characterized in that said seventh sub-procedure (980) comprises the step of assigning each pair of virtually distinct destination sites obtained following bisection to the respective pair of logically contiguous outputs respecting the assignment criterion with where the place of destination from which this pair of virtually distinct places of destination originates following bisection is assigned to the respective subset of outputs in the last sorting cycle of the sequencing process. 11.- Method according to any one of claims 2, 3, 4, characterized in that it further comprises the steps of: - checking (190) whether the improvement obtained with said first sub-procedure (100) is insignificant or whether a first predetermined maximum time (tO) has elapsed since the first sub-procedure (100) started; And - repeating said first sub-procedure (100) if the improvement obtained is significant or said first predetermined maximum time (tO) has not elapsed. And - carrying out said second and third sub-procedures (210-230, 240) if the improvement obtained is not significant or said first predetermined maximum time (tO) has elapsed. 12.- Method according to Claim 11, characterized in that it further comprises the steps of: - checking (200) if there are outputs (U) of the machine (1) available to be assigned to said subsets (A, B, C, D); - carrying out said second sub-procedure (210-230) if there are available outputs to be assigned to said subsets (A, B, C, D); And - carrying out said third sub-procedure (240) if there are no outputs available to assign to said subsets (A, B, C, D). 13.- Method according to Claim 12, characterized in that it further comprises the steps of: repeating said first sub-procedure (100) after carrying out said second sub-procedure (210-230). . 14.- Method according to any one of claims 11 to 13, characterized in that it further comprises the steps of: - checking (250) whether the improvement obtained with said third sub-procedure (240) is not significant or whether a second predetermined maximum time (tl) has elapsed since the second sub-procedure (240) started; - repeating said third sub-procedure (240) if the improvement obtained is significant or said second predetermined maximum time (tl) has not elapsed; And - carry out said second procedure (300, 530, 730, 980) to improve the load balancing of the outputs (U) of the machine (1) at the end of the first sorting cycle if the improvement obtained is not significant or said second time has elapsed fixed maximum (tl). 15.- Method according to any one of claims 6, 7, 8, 9, characterized in that it further comprises the steps of: - checking (720) whether the improvement obtained with said fifth sub-procedure (530) is not significant or whether a third predetermined maximum time (t2) has elapsed since the fifth sub-procedure (530) started; And - repeat said fifth procedure (530) if the improvement obtained is significant or said third predetermined maximum time (t2) has not elapsed; and - carrying out said sixth sub-procedure (730) if the improvement obtained is not significant or said third predetermined maximum time (t2) has elapsed. 16.- Method according to Claim 15, characterized in that it further comprises the steps of: - checking (970) whether the improvement obtained with said sixth sub-procedure (730) is not significant or whether a fourth predetermined maximum time (t3) has elapsed since the sixth sub-procedure (730) started; And - repeat said sixth procedure (730) if the improvement obtained is significant or said fourth predetermined maximum time has not elapsed (13); and carrying out said seventh sub-procedure (980) if the improvement obtained is not significant or said fourth predetermined maximum time has elapsed (13). 17. Method according to claim 16, characterized in that it further comprises the steps of: - check (990) whether the improvement obtained with said seventh sub-procedure (980) is not significant or a fifth predetermined maximum time (14) has elapsed since the seventh sub-procedure (980) started; And - repeat said seventh procedure (980) if the improvement obtained is significant or said fifth predetermined maximum time (t4) has not elapsed; and terminating said sequencing process if the improvement obtained is not significant or said fifth predetermined maximum time <14) has elapsed. 18. Method according to any one of the preceding claims, characterized in that it comprises the step of: executing said fifth sub-procedure (530) after said fourth sub-procedure (300). 19.- Method according to any one of the preceding claims, characterized in that it further comprises the step of: perform, before said first and second procedures for improving the load balancing of the machine outputs (1) at the end of the second and, respectively, of the first sorting cycle, a third procedure (30-80) to verify the sufficiency of resources for carrying out the sequencing process; - carry out said first and second procedures (100, 210-240, 300, 530, 730, 980) to improve the load balancing of the outputs (U) of the machine (1) at the end of the second and, respectively, of the first cycle sorting if the third procedure (30 - 80) has a positive outcome; And - if said third procedure (30-80) has given a negative result, terminate said sequencing process and report which insufficient resources have caused the sequencing process not to be executed. 20.- Method of load balancing in an object sequencing process, in particular postal objects, substantially as described with reference to the attached drawings.