WO2003019427A1

WO2003019427A1 - A method and arrangement for handling information

Info

Publication number: WO2003019427A1
Application number: PCT/AU2002/001080
Authority: WO
Inventors: Patrick Conrick
Original assignee: Patrick Conrick
Priority date: 2001-08-28
Filing date: 2002-08-13
Publication date: 2003-03-06
Also published as: AUPR727101A0

Abstract

An interaction flow matrix (10) lists the source and sink elements of a process on the axes of a matrix. The interactions are entered in the nodes of the matrix according to a chosen convention to indicate the direction of the flow of the interaction. A visual indication of the priority of the status of the interactions is provided by the use of visual coding. The nodes of the matrix can be electronically linked to a tabular explanation of the interactions (50). A pyramid matrix (31, 32, 33), can be used to provide one or more levels of supervision. The matrix may be applied in the control of a process (71) in which monitors (73) update the matrix (74) and wherein the process is controlled by controllers (72) in response to the matrix information.

Description

A method and arrangement for handling information

Technical field

This invention relates to a system and method of analyzing or controlling a collection of data or the interactions of a process. The invention has the advantage of being able to make large amounts of information available in a readily accessible manner by displaying the information in an array or matrix form. In a particular embodiment, the invention provides a matrix as a process analysis or management tool. It may be used, for example, as a tool for analyzing, managing or controlling business or industrial processes, organization structures, interactions, or information flows.

A particular application of the invention uses a multi-layered matrix.

Background art

In Innovation at the Speed of Information, Harvard Business Review OnPoint , January 2001. Steven D. Eppinger discusses a Design Structure Matrix (DSM). In the DSM, the actions are listed chronologically on both axes. The matrix can be used to identify and reduce feedback by changing the order of the actions. This publication does not disclose a technique for providing an indication of the direction of the interaction flow between the parties or elements of the process. Nor does it suggest the use of the matrix to indicate the status of an action.

Karl Eric Sveiby in his book "The New Organizational Wealth", at page 47, discloses a 2*2 matrix with TO and FROM axes converting between tacit and explicit knowledge through different processes at the nodes. Tthe publication does not suggest that this matrix can be used as a process monitoring or control tool. Nor does it suggest the use of the matrix to indicate the status of an action.

Disclosure of the invention

It is desirable to provide a data structure which can be used to make information available in a readily accessible form. It is also desirable to provide a method and means to indicate the status of an interaction.

Preferably, the data structure is adapted to make large amounts of information available in a condensed form, with ready access to fuller details.

Preferably, the data structure can be used to indicate interaction flows in a process.

According to a first embodiment of the invention, there is provided a method of representing, monitoring, or controlling interactions between one or more source elements and one or more sink elements, the method including: identifying interactions between pairs of source/sink elements; establishing a flow direction convention indicating the direction of flow of interactions between pairs of elements from source elements to sink elements; listing the source elements on a first axis of a matrix of nodes in accordance with the flow convention; and listing the sink elements on a second axis of the matrix of nodes in accordance with the flow convention.

A further embodiment of the invention includes inserting an interaction status flag in one or more of the nodes.

In a further embodiment, the method includes the step of inserting an interaction identifier at the intersection node between the or each pair of source/sink elements in accordance with the flow direction convention.

According to a preferred embodiment, the method also includes the steps of monitoring one or more of the interactions, and changing the corresponding node or nodes to indicate changes in the status of the monitored node or nodes. Preferably, a change in status of an interaction is indicated by a change in colour of the corresponding node. Alternatively or additionally, a change in status is indicated by changing a number in the corresponding node.

In a further embodiment one or more additional row and/or column sub-nodes are added as required for the or each element to indicate multiple interactions between the element or elements provided with one or more additional sub-nodes and any element.

In a further embodiment an interaction table is provided, the nodes contain references which are included in the table associated with a descriptor of the interaction.

In a further embodiment the matrix and the table are provided in electronic form, and each nodes is electronically linked to a corresponding descriptor in the table.

In a further embodiment the invention provides a method of representing a plural layer structure having at least one subordinate layer and an apex array, the method including providing one or more matrixes at each subordinate layer, and progressively concentrating the subordinate layers into the apex array.

In a further modification of this embodiment, the highest priority indicator is carried forward from each inferior layer, so that the apex array includes a record of the highest priority nodes in the inferior layer matrixes.

The invention is also embodied in an interaction flow matrix indicating the flow of interactions between a plurality of elements, the matrix including: a listing of elements on both the vertical and horizontal axes of a matrix of nodes; and interaction identifiers located at the intersection node between a pair of nodes in accordance with a flow direction convention indicating the direction of flow of interactions between pairs of elements corresponding to the or each node. In a further embodiment of this matrix, one or more additional row and/or column nodes are added as required for the or each element to indicate multiple interactions between the element or elements provided with one or more additional sub-nodes and any element.

Preferably the nodes are coded according to a visual priority code to provide a status indication for the corresponding interaction.

In a further embodiment of the matrix an interaction table is provided, wherein the nodes contain references which are included in the table associated with a descriptor of the interaction.

Preferably the matrix and the table are provided in electronic form, and the nodes are linked to the corresponding item in the table.

A particularly useful embodiment of the invention provides a pyramid matrix representing a plural layer structure, the pyramid matrix including one or more matrixes at each inferior layer, the inferior layers being progressively concentrated into an apex matrix.

Accordingly, an embodiment of the invention includes a pyramid matrix arrangement including an hierarchical arrangement of arrays of nodes, the arrangement including at least one base level array of nodes, the base level array of nodes including a plurality of base nodes organized into one or more groups of nodes, and one or more higher level nodes, each lower group of nodes being linked to at least one higher level node to transfer information about the contents of the base level array of nodes to the or each corresponding linked higher level node.

Preferably, the above arrangement includes two or more successively higher level arrays of nodes, wherein each group of nodes in each level is linked to at least one node in the next highest level array up to the highest level array. In one embodiment there is at least one node in at (east one intermediate layer whose contents originates at that intermediate level.

A further embodiment of the invention provides a method of transferring information about a base level array of nodes including : monitoring the base level array, determining at least one characteristics of information in the base level array, and reproducing the thus determined characteristic in the linked higher level node.

An embodiment of the invention also provides a method of monitoring a process including: monitoring one or more steps of the process; associating the result of monitoring each step in a corresponding node of a base level array of nodes; setting a visual flag in the corresponding node in accordance with the result of the monitoring the step.

An further embodiment of the invention also provides, in a pyramid matrix arrangement, a monitoring method including organizing the nodes at each level into one or more groups of nodes; at each level except the highest level, linking each group of nodes to at least one node of a next higher order array of nodes; monitoring the information in each group of nodes; and setting a visual flag in the linked higher level node in accordance with a characteristic derived from the linked group of nodes.

Brief description of the drawings

The invention will be described with reference to the embodiments shown in the accompanying drawings, in which:

Figure 1 illustrates the structure of a first embodiment of an interaction flow matrix; Figure 2 illustrates a coding process for use in an embodiment of the invention; Figure 3 illustrates a pyramid matrix according to an embodiment of the invention; Figure 4 shows a spreadsheet layout suitable for implementing the matrix; Figure 5 shows a table illustrating the explanations of the node contents of a matπx embodying the invention;

Figure 6 shows an embodiment of a process control using a two level matrix and associated control terminals;

Figure 7 shows a pyramid matrix with three levels;

Figure 8 shows a flow chart for different levels a monitoring process;

Figure 9 shows a communication network structure implemented as a multi-layer matrix arrangement.

Description of the invention

This invention may be embodied in an Interaction Flow Matrix (IFM) which may be used as a tool for analyzing, managing or controlling data structures or the interactions between the various elements of a process. The specification discusses the application of the matrix to a process, for example, within an organization. In the IFM the elements or parties to a process are listed on the axes of the matrix, and the nodes represent the interactions. In some cases, all the elements are listed on both axes of the matrix. A flow convention may be assigned to the IFM to provide additional information for the analysis or control of the process. One or more row elements and/or one or more column elements may be associated with one or more sub-rows or sub-columns respectively.

The invention utilizes the concept of data flows or interaction flows between elements or parties in a process, and provides a graphical technique to represent the flows/interactions and their direction at the intersection nodes on the chart.

The Interaction Flow Matrix differs from the DSM in that, in the DSM, the actions are listed on the axes, whereas, in the IFM, the elements or actors in the process are listed on the axes, and the interactions are listed or represented at the intersections, enabling the direction of the flow of the interaction to be indicated on the matrix. The DSM is used to optimize the sequence of the actions in a static, pre-operational process. The IFM structure facilitates analysis of the interactions by identifying the source and sink of each interaction, and can be used in real-time processes. The interaction flow matrix 10 as shown in the embodiment of Figure 1 is deployed by first identifying the elements of a data structure, interaction or process which it is desired to analyze. These elements are then written on both axes of a matrix. With reference to Figure 1, a simplified interaction matrix is illustrated. A convention has been adopted to indicate the direction of the interaction flow between the elements in which the direction of flow is from the elements listed in the top row to the elements listed in the left column. This interaction matrix contains four elements, A, B, C, D, and these elements are listed in both the top row, under the heading FROM, and the left column, headed TO. The arrow illustrates the direction of flow, or the source and sink of the interaction.

The matrix has the virtue that, because the elements are listed on both axes, any interaction between any two elements can be illustrated at the appropriate intersection node. Thus the node containing the label BA indicates an interaction originating at B and flowing to A. Conversely, the node labeled AB indicates an interaction originating at A and flowing to B. That is, this chart adopts the convention that the flow is from the column label element to the row label element. It is clearly within the scope of the invention for the converse convention, in which the flow is from the row label element to the column label element, to be adopted. Similarly, it is a matter of choice as to whether the labels are at the top or the bottom, or to the left or the right.

Bi-directional flows are represented by the corresponding pair of diagonal mirror image nodes , eg, AB - BA; CD - DC; etc.

The intersection nodes in Figure 1 are labeled with the letters identifying the intersection. However, in one embodiment, the nature of the interaction may be directly inscribed in the interaction node. Alternatively, in an embodiment shown in Figure 5, a table of interactions may be compiled, related to the matrix by the node references. In a preferred embodiment, the matrix and table may be embodied in electronic form and linked, eg, by the use of HTML. Thus a node in a matrix may be linked to a table or an item in the table by HTML, so that a user can click on an item in a node of the matrix and have access to an explanation or fuller details of the item indicated in the node. Uses of the interaction matrix include the analysis or design of a process, or the monitoring or control of a process.

Figure 1 also indicates another feature of the matrix, the ability to include more than one interaction between any two elements. To do this, Figure 1 provides additional rows for the additional interaction. As illustrated in Figure 1 , there are two A rows, so that two interactions can be illustrated between any element and the A element. This technique is suitable where an element is the recipient of multiple inputs from one or more elements. Where an element is the source of several flows to one or more elements, this can be accommodated by providing two or more columns for the element which is the source of multiple flows. Thus it is envisaged that one or more row elements may have more than one row associated with it, and one or more column elements may have more than one column associated with it. In other words, sub-nodes are used to indicate multiple interactions between two elements.

Where three or more elements participate in an interaction, this is indicated by labeling each node corresponding to an interaction between a source and a sink involved in the interaction. This makes it possible to identify each of the specific pairs of elements involved in the interaction, and to monitor the corresponding nodes.

An enhancement of the matrix includes the use of a colour code to indicate the effectiveness of the various interactions of a process.

Figure 2 illustrates a colour coded matrix. For example, the red, amber, green code may be used to indicate whether an interaction is not working, in need of improvement, or satisfactory. This makes the matrix a useful management tool which gives a quick overview of the process which clearly indicates the problem areas in a process. The responsible personnel investigating the effectiveness of a process allocate the priority code on the basis of analysis of the process. In a particularly useful embodiment, the matrix may be used for real time monitoring or control of a process, in which the cells are controlled by the output of monitors which monitor interactions in the process and control the status indication of the nodes accordingly. Thus in Figure 2, the nodes CB and CD are coloured yellow, indicating a need to improve these interactions, while node BC is coloured red indicating a breakdown of the interaction, and immediate remedial action is required.

The visual coding can also be used to map critical paths in a process.

Figure 2 illustrates a visual coding process for use in an embodiment of the invention.

In Figure 2 most of the interactions (green) are working well, but the interaction from B to C is not working, and the flow from C to B is in need of improvement. This may lead to, for example, the interaction flowing from C and D also being affected if this interaction requires the BC interaction to work. A failure in one interaction may cause the total failure of the process, depending on the degree of interconnectivity between the interactions and the criticality of the defective interaction. The matrix provides a means of analyzing complex multi-element processes and structures, and providing an indication of the status or effectiveness of the elements of the process.

Figure 3 illustrates a particularly useful application of the matrix in analyzing multilayer processes or organizations. In a multi-layer process or organization, in which the lower layers are reflected in the higher layers, the problems in the lower layers can be made apparent at the higher levels. This is done by the use of a matrix pyramid in which multiple lower layers are concentrated into a higher level of matrixes (matrices), up to a single overview layer. To maintain visibility of problems at the lower layers, the higher Jevel node reference to the lower level is coloured as the highest priority of the lower level process to which it is linked. In Figure 3, the overview layer 31 collates information from intermediate layer 32. Preferably, there is at least one cell in the overview layer matrix 32 linked to an intermediate matrix in the intermediate layer 32. The cells in the overview layer 31 are programmed to indicate the most urgent or highest priority status from the corresponding linked matrix in the intermediate layer 32. Similarly, each cell in each intermediate layer matrix is linked to a matrix at the base or functional layer 33, and indicates the highest priority or most urgent status from the corresponding matrix of the functional layer. Thus, the highest priority status from each matrix or group of matrixes in the functional layer is linked through to the corresponding node in the intermediate layer 32 and hence to the corresponding node in the overview layer 31.

In the pyramid matrix, each process matrix at the lower level is represented by at least one node of the next higher level matrix with which it is associated. Thus one or more of the nodes in one or more of the higher level matrixes represent a corresponding lower level matrix. There may also be other nodes in the higher level matrixes which represent interactions which originate at that higher level.

In one embodiment two or more nodes of a higher level matrix may be used to report an associated lower level matrix, allowing for the reporting of additional information from the lower matrix.

In a preferred embodiment, the matrix is implemented in an electronic spreadsheet such as illustrated in Figure 4. The nodes of the matrix are formed by the intersection cells of the spreadsheet.

Where there are large numbers of elements, it is possible to take advantage of the ability of spreadsheets such as LOTUS 123, EXCEL etc. to hide columns and rows while leaving the header row and column visible when looking at the flows from specific elements. Using a database to store large amounts of operational detail is also accommodated by the invention. Specific views can be designed for the analysis of individual processes or structures in a company. Preferably, these views present the information in a matrix format with source and destination axes. This facility may be important when it is considered that there are potentially thousands of elements at the functional or base level, allowing for the possibility of two or more interactions between each element pair. Thus the ability to design specific views to look in detail at specific processes or interactions between less than all elements makes accessibility and comprehensibility of the sections of an organization or process.

Another way of compressing the matrix is to create an interaction table and assign codes to the various interactions, so that the code can be inserted in the intersection nodes. This makes it possible to make an overall presentation highlighting the interactions which are not working or which are in need of improvement on a single consolidated chart.

In Figure 3, a "pyramid matrix" representing an organization with a plurality, m, of functional processes, eg, manufacturing processes, is shown. Only the matrix for process n is shown for clarity at the lower level, but the invention envisages more than one matrix at this level and at the intermediate level. Each matrix at the lower level is associated with one or more cells of at least one matrix at the next higher level. One of the nodes of this matrix n is red and another is amber. The cell in the next higher level matrix which is linked to the process matrix n is thus coloured red, which is the highest priority level in the n matrix. Similarly, the cell in the overview matrix to which this intermediate level matrix is linked is shown as red. The levels of matrixes may mirror the company reporting structure. This makes it possible to readily identify problem areas immediately from an inspection of the overview matrix.

Given that it is possible to obtain several orders of magnitude of concentration at each level, very large processes can be monitored at the top level with a simple display of, eg, ten cells, while it is possible to drill down to the bottom level in one less steps the number of levels. For example, if the top level has 10 cells, each cell being associated with a 10*10 intermediate matrix, there are thus 1000 cells at the intermediate level. Again, each intermediate level cell may be associated with, eg, a 10*10 matrix at the base level. Thus, there may be 100,000 cells at the base level which are accessible in two steps from the top level, while requiring only 10 cells at the top level to provide a monitoring function for all the 100,000 cells at the base level.

Again, if the pyramid matrix is in electronic form, the nodes can be linked electronically so it becomes a simple matter to track down to the source of a problem from the overview layer, for example using hyperlinking. Ideally, the overview matrix could be a single row, multiple column matrix of coloured nodes indicating the performance of the underlying business divisions to the company executive. Clicking on one of the nodes of the row gives access to, for example, the underlying business division, and clicking on nodes in the business division gives access to the corresponding underlying product line or business unit matrix. Knowledge management is an evolving concept, and, while there are several tools for data warehousing and data retrieval, this invention may be used to provide organizations with a tool for analyzing the information flows in which the organization is involved to provide a basis for better organizing those processes.

The interaction matrix can be utilized to analyze multi-element processes. One use to which the matrix can be put is to analyze the efficiency of a process, particularly in eliminating sequential operations which could be performed in parallel. In other words, where an interaction flows from B to C and then from C to A without input from C, this could be improved by having simultaneous flows from B to C and B to A. The degree of detail included in the interaction matrix must be sufficient to carry out this analysis. Different degrees of detail may be included in the matrix, depending on the purpose to which the matrix is to be put.

in some processes, there may be some elements which are sources only, and some elements which are sinks only, so that it is not necessary to include these elements on both axes. This fact can be used to reduce the size of the matrix. If there are p elements which are sources only, q elements which are both sources and sinks, and r elements which sinks only, this would produce a (p + q)*(q + r) matrix, rather then the (p + q +r)*(p + q + r) matrix if all elements were included on both axes. Thus a single row, multi column-matrix or list is within the concept of the invention, as is a single column, multi-row matrix or list.

An example of pure sources and sinks is a grocery store, where the suppliers are sources only, and their products sinks only, ie, the suppliers deliver the products to the store, the delivery operation being the interaction. Such a matrix would have the suppliers listed only on the source axis and the products listed only on the sinks axis.

Figure 6 shows a simplified supermarket store operation with a delivery system 601, a temporary store 602, and display shelving 603. A goods inward monitor 604, shelf monitor 605, and checkout monitor 606 monitor the goods into the temporary store, the transfer of goods to the display shelves, and the sale of products at the checkout. Thus the goods are monitored on delivery, and on being placed on the shelves and at the point of sale. The monitors are connected to a processor 607 which controls the product matrix. In this example there is a plurality of different product matrixes divided by the nature of the goods, and a team would be allocated to each group of goods to ensure the goods are available for purchase. Alternatively, the goods may be delivered automatically to the shelves where an automatic delivery system is provided.

Figure 6 shows matrix screens for CEREAL 608, DAIRY 609, and UTENSILS 610 by way of example. Optionally, the screens for the shelves are arranged so that the position of the nodes on the screen correspond with the position of the corresponding physical products on the shelves. Thus, as shown in Figure 6, it is the product in the second row of the second bank of shelves which has triggered the action.

The monitors 604, 605, 606 are connected to a processor 607 which can thus determine the number of goods in the back store and on the shelves. The processor is programmed with status thresholds, depending, for example, on the turnover of the specific goods, so that when the number of a particular type of product falls below a first threshold, the matrix cell corresponding to that product is switched from GREEN to YELLOW. This is a notice that the product shelves should be restocked soon. If the number of that product falls below a second threshold this indicates that the product shelves must be restocked immediately. The system can also monitor the back room store in a similar manner.

In the present example, one of the DAIRY products on the DAIRY SCREEN 609 has triggered a threshold action and this is indicated in Figure 6 by the cross-hatched node, but in reality could be a YELLOW or RED highlighting of the node. The nodes of the product screens relate to specific products from particular suppliers. This information is available to the DAIRY team who can restock the shelves as appropriate. The restocking is recorded by way of the shelf monitor 605, so the system automatically resets when the shelves are restocked to the prescribed level.

Such the pyramid matrix for such an arrangement can be implemented in Microsoft EXCEL by defining one or more base level matrixes corresponding, for example, to the product/supplier groups the functional team level of Figure 6. The nodes are programmed to use conditional formatting in which, for example, the nodes are GREEN when the node content <2, YELLOW when =2, and RED when>2. Thus, the controller writes either 1, 2, or 3 in the node depending on the indicated status, and the colour of the node is changed appropriately.

To refer the highest priority node up to the corresponding node of the supervision matrix, a formula is used to identify the MAXIMUM value in the corresponding base level matrix. This may be done on the same sheet as the base matrix or directly into the corresponding node of the supervision matrix. Again the nodes of the suprevision matrix are conditionally formatted to reproduce the colour flags GREEN, YELLOW, or RED. By using the formula to determine the maximum value within the range of the base matrix, the priority of the highest priority node is transferred to the node of the suprevision matrix which represents the corresponding base level matrix.

The threshold trigger is mirrored in the supervision level screen 612, so a supervisor can monitor the status of all the groups of products from a single display.

Figure 7 shows a multi-layer monitor and control system. In this embodiment, three layers are shown, but the invention encompasses from one layer upwards. A process is illustrated figuratively by arrow 71. A plurality of monitors 72 monitor different actions along the process and feed this to the matrix processor and 74. Functional level operation and control is provided at terminal 75, which also is connected to an associated one of a plurality of controllers 72 to enable the process to be controlled in response to the matrix displaying the monitor conditions. There may be, eg, one functional terminal 75 for each functional level node, or each functional terminal may operate several node controls.

An intermediate layer of supervision terminals 76, is provided which concentrates two or more functional level nodes from the base matrix. This may, where desired, have override control of the corresponding controllers. The top layer may include one or more terminals 77. The concentration of nodes provided by the cascaded matrixes is represented conceptually by the pyramid shape.

Figure 8A is a flow chart of the process of monitoring the base level functions. The base functions are monitored against the selected threshold. When the threshold is broken, a check is made as to the level of the threshold which is breached. For example a decision is made as to whether the priority is RED or YELLOW, and the node is flagged accordingly. In the case where an error report table is provided, than error message is also written to the error report table, explaining the nature of the problem. The status flag is then transferred up to the or each higher level.

Figure 8B shows a flow chart for the top level matrix process. When an alert flag is detected, the user can drill down through the layers to find the source and nature of the problem.

Another way of using the matrix is to include only the sources and destinations involved in any specific action in the row and column axes. The results is a simplified matrix in which it is easier to study the specific action. For this level of analysis it may be useful to use the highest level of detail. That is, the highest level of detail can be used to analyze specific actions at the fundamental level. This level of detail may correspond to the level of process designer or process operator.

One particular application of the invention is in the real-time reporting of a company's performance. Where a company has a fully integrated computerized system monitoring the operations of the business, critical measures may be constantly monitored. For example, the on-time delivery of products may be monitored against promised delivery dates. Delays in production having flow-on consequences for delivery may also be monitored so that timely remedial action can be taken. The delays are relayed up to the senior executive's monitoring screen and can be analyzed by the senior executive simply by drilling down to the source of the delay. The executive monitor display need only be a simple row of coloured nodes with an associated row of business division labels. The nodes should always be green. In the event of a code yellow in any of the nodes, the executive can immediately trace this to its source and find an explanation for the problem from a linked associated explanatory table, or can contact the appropriate manager to discuss the problem and the necessary action.

Figure 9 is an example of the use of the pyramid matrix to implement the connections in a communication network, for example an ARM network. While it is currently fashionable in telecommunications to refer to a "domain", we will maintain reference to a "matrix" for conistency within the specification.

As shown in Figure 9, the matrix can be used to model telecommunication networks. For example, the base level matrixes may represent local networks grouped into sets of local networks to which individual customers may be connected, the intermediate layer may represent the interconnexions between the local networks in designated groups, and the overview layer may represent the interconnexions between intermediate layer matrixes. Clearly, the arrangement is expandable to include further levels which can progressively condense the traffic from subordinate layers. It is usually the case that the higher up the levels, the greater is the bandwidth (traffic carrying capacity) of the links.

Individual customers connected to the same matrix require only local level connexions, ie, intra-matrix connexions, while customers connected to matrixes in the same group of matrixes require one additional degree of connection between the matrixes of that group, ie, inter-matrix connexions, analogous to connexions within a local exchange compared with connexions between adjacent local exchanges - (this latter is often done now by going to a higher level and returning to the other local exchange rather than by direct connexion between the local exchanges). Where a customer needs to be connected to a customer of a matrix of another group of matrixes, the connexion must be made via the next higher level if the groups of matrixes involved are covered by the same intermediate level matrix. If this is not the case, the connexion is referred to the next highest level, and so on.

However, the required amount of connexion control may be reduced by providing multiple connections to nodes at the intermediate levels from the lower levels. This is shown in Figure 9, where groups of matrixes at the lower level are linked to two or more nodes of the intermediate level. Thus the group 901 in Figure 9A connects to the four nodes 902 at the base level at the intermediate level. Similarly, the group 921 at the base level in Figure 9B connects to the four nodes 922 at the intermediate level. There is an overlap of two nodes between the groups of nodes 902 and 922 at the intermediate level. Thus there is the possibility to make an intra-node connexion between the base groups 901 and 921. The pattern of connexion between the lower matrixes to nodes of the intermediate matrix can optimaly be arranged to provide overlap between the connexions at the intermediate level, such that each lower matrix shares at least one intermediate node with each of the other matrixes of the same lower matrix group. This reduces the inter-matrix switching at the intermediate level as the connexions can be made at the intra-matrix level in the intermediate level matrixes instead of at the inter-matrix level.

The trade off is that additional communication links (optical, wireless, wired) are required between matrixes, so that each matrix at the lower level is connected to two or more nodes at the next higher intermediate level. It is often not practical to provide full redundancy and overlapping between all the matrixes. However, where the traffic flows justify the expense, this technique may prove an effective method of reducing the control processing load on the network.

Ideally the intermediate layer for each set of base layer matrixes should have one common node for each pair to reduce the number of links required to set up a communication between a pair of nodes in two different matrixes within the same group. In Figure 9A, the base level matrix 901 is associated with, eg, three intermediate level nodes 902 as shown, the base matrix 921 is associated with four intermediate layer nodes 922 including two nodes shared with 902. This arrangement may be optimized to provide a fault tolerant arrangement balanced against the use of network resources.

Similar considerations may be applied to the upper level, eg, 903, 923. Common nodes may be associated with each pair of intermediate matrixes to provide efficient communication within the common upper level node, while another independent node may be associated with each respective intermediate level matrix to provide a degree of fault tolerance. Thus with three intermediate level matrixes, there are three nodes associated with each intermediate level matrix. Of the three associated with each intermediate level matrix, two are associated with one or other of the other two intermediate matrixes, while the third upper level node is not associated with the other two matrixes. Thus it is possible to make a direct inter-matrix connection at the upper level between any pair of intermediate level matrixes through the associated common node. However, if the common node fails or is overloaded, it is still possible to establish the connection through either of the other two nodes associated with the first intermediate matrix and the other two nodes associated with the other intermediate matrix involved in the communication.

Claims

Claims:

1. A method of representing, monitoring, or controlling interactions between one or more source elements and one or more sink elements, the method including: identifying interactions between pairs of source/sink elements; establishing a flow direction convention indicating the direction of flow of interactions between pairs of elements from source elements to sink elements; listing the source elements on a first axis of a matrix of nodes in accordance with the flow convention; listing the sink elements on a second axis of the matrix of nodes in accordance with the flow convention; inserting an interaction status flag in one or more of the nodes.

2. A method as claimed in claim 1 , including the step of inserting an interaction identifier at the intersection node between the or each pair of source/sink elements in accordance with the flow direction convention.

3. A method as claimed in claim 1 or claim 2, including the steps of monitoring one or more of the interactions and changing a characteristic of the corresponding node or nodes to indicate a change in the status of the monitored interaction.

4. A method as claimed in claim 3, including the step of indicating a change in status of an interaction by a change in colour of the corresponding node.

5. A method as claimed in claim 3 or claim 4, including the step of alternatively or additionally indicating a change in status of an intercation by changing a number in the corresponding node.

6. A method as claimed in any one of claims 1 to 5, wherein one or more additional row and/or column nodes are added as required for the or each element to indicate multiple interactions between the element or elements provided with one or more additional sub-nodes and any element.

7. A method as claimed in any one of claims 1 to 6, including providing an interaction table, wherein the nodes contain references which are included in the table associated with a descriptor of the interaction.

8. A method as claimed in claim 7, wherein the matrix and the table are provided in electronic form, and wherein each node is electronically linked to a corresponding descriptor in the table.

9. A method of representing interaction flows substantially as herein described with reference to the accompanying drawings.

10. A method of presenting information in a plural layer structure of nodes having at least one subordinate layer and an apex layer, the method including providing one or more matrixes as claimed in any one of claims 1 to 9 at each subordinate layer, and progressively concentrating the subordinate layers into the apex layer.

11. A method as claimed in claim 10, including the step of carrying forward a highest priority indicator from each inferior layer to the next highest layer, so that the apex layer includes a record of the highest priority nodes in the inferior layer matrixes.

12. An interaction flow matrix indicating the flow of interactions between a plurality of elements, the matrix including a listing of the elements on both the vertical and horizontal axes of a matrix of nodes; interaction identifiers located at the intersection node between a pair of nodes in accordance with a flow direction convention indicating the direction of flow of interactions between pairs of elements.

13. A matrix as claimed in claim 12, wherein one or more additional row and/or column nodes are added as required for the or each element to indicate multiple interactions between the element or elements provided with one or more additional sub-nodes and any element.

14. A matrix as claimed in claim 12 or claim 13, wherein interactions are coded according to a visual priority code.

15. A matrix as claimed in any one of claims 12 to 14, wherein an interaction table is provided, and wherein the nodes contain references which are included in the table associated with a descriptor of the interaction.

16. A matrix as claimed in claim 15, wherein the matrix and the table are provided in electronic form, and wherein the nodes are linked to the corresponding item in the table.

17. A pyramid matrix presenting information in a plural layer structure, the pyramid matrix including one or more matrixes as claimed in any one of claims 12 to 16 at each inferior layer, the inferior layers being progressively concentrated into an apex layer

18. A matrix substantially as herein described with reference to the accompanying drawings.

19. A pyramid matrix substantially as herein described with reference to the accompanying drawings.

20. A pyramid matrix arrangement including an hierarchical arrangement of arrays of nodes, the arrangement including at least one base level array of nodes, the base level array of nodes including a plurality of base nodes organized into one or more groups of nodes, and one or more higher level nodes, each lower group of nodes being linked to at least one higher level node to transfer information about the contents of the base level array of nodes to the or each corresponding linked higher level node.

21. An arrangement as claimed in claim 20, including two or more successively higher level arrays of nodes, wherein each group of nodes in each level is linked to at least one node in the next highest level array up to the highest level array.

22. An arrangement as claimed in claim 20 or claim 21 , including at least one node in at least one intermediate layer whose contents originates at that intermediate layer level.

23. A method of transferring information about a base level array of nodes including : monitoring the base level array, determining at least one characteristics of information in the base level array, and reproducing the thus determined characteristic in the linked higher level node.

24. A method of monitoring a process including: monitoring one or more steps of the process; associating the result of monitoring each step in a corresponding node of a base level array of nodes; setting a visual flag in the corresponding node in accordance with the result of the monitoring the step.

25. In a pyramid matrix arrangement as claimed in any one of claims 20 to 22, a monitoring method including, organizing the nodes at each level into one or more groups of nodes; at each level except the highest level, linking each group of nodes to at least one node of a next higher order array of nodes; monitoring the information in each group of nodes; and setting a visual flag in the linked higher level node in accordance with a characteristic derived from the linked group of nodes.

26. A network arrangement including: a base level including one or more base level matrixes, the or each base level matrix including one or more nodes to which customers are connected, and one or more higher levels, each level including one or matrixes of one or more nodes, the highest level including at least one matrix having at least one node, wherein each matrix at a lower level is linked with at least one node of an associated matrix at the next highest level, and wherein at least two nodes at one level are linked with at least two nodes at the next higher level, one of the said two nodes at the next higher level being linked to both of said at least two nodes at one level.

27. An arrangement as claimed in claim 26 wherein, at each level, the matrixes are formed into matrix groups, and wherein, at each higher level, the immediately lower matrixes are linked with a common node for each other matrix in a same group, whereby any matrix at the lower level can establish a connection to any other matrix at that lower level via a single node at the said next higher level.

28. A network arrangement substantially as herein described with reference to the accompanying drawings.