WO2003052646A1

WO2003052646A1 - State based management for queue-server systems

Info

Publication number: WO2003052646A1
Application number: PCT/AU2002/001711
Authority: WO
Inventors: Margaret Roslyn Rossiter
Original assignee: Margaret Roslyn Rossiter
Priority date: 2001-12-18
Filing date: 2002-12-17
Publication date: 2003-06-26
Also published as: AUPR954701A0

Abstract

A method is disclosed of managing an over-the-counter queue-server system in which a plurality of queuers wait for a service to be performed by at least one server, the method including: each operating server ascertaining the nature of a service to be provided, and prior to or soon after commencing the service for a queuer, providing an indicative estimate of the time required for that server to perform the service, the indicative estimate being based on the individual server's knowledge of his or her operating ability and the type of service requested by the queuer, and initiating a service response if the estimate, or if more than one server at least a predetermined number of the estimates, exceeds a predetermined minimum time.

Description

"State Based Management for Queue-Server Systems"

Technical field

This invention relates to a system for, and method of, queue-server system management, and in particular to over-the-counter (OTC) services such as are typically provided in commercial face to face situations.

The technique of the present invention will be seen to be applicable in any server system where customers with varying demands form at least one queue. Such services could also be provided at remote locations and that the customers seeking the service do not necessarily queue in the physical sense of standing in line. Some specific areas, by way of non-exhaustive examples of queueing, include:

■ Banking Chambers - OTC service - a single first-in first-out queue forms before a number of tellers.

■ Government Shop Fronts - OTC service * Health Insurance Offices - OTC service

• Call Centres - single queue of phone customers and a number of service operators.

■ Repair/Service operators - eg. Service or repair calls from telecommunication customers (a queue) to be handled by a number of service technicians.

■ The provision of maintenance and repair services to a vehicle fleet. ■ The availability of a bed or an operating theatre in a hospital, or the availability of triaging facilities at hospital emergency and outpatient reception.

■ The internet coordination of services provided across the internet by dispersed service providers - such as for example, home page design.

For illustrative purposes the invention will be described with reference to OTC banking services. However it will be appreciated that the invention does not have exclusive application to such services and can be used in any situation where a user has to queue for service (in the sense that the expression "queue" is used herein).

As will subsequently become apparent, the present invention centres on what is termed a game-play methodology, and which has been developed from analysis of team sport tactics. The game-play methodology is a process improvement methodology for human-activity systems - targeting specifically the application of constraints. The game-play methodology is effective in enhancing the performance of a team whether in sporting pursuits or when applied to business, commercial, industrial or like pursuits. Accordingly, at its broadest level the invention also relates to enhancing team performance by a game-play methodology. As will subsequently become apparent, the present invention is a form of game-play that is applicable in queue-server systems and is termed State Based Management (SBM). In business a game-play is a set of constraints applied to the members of a team with the intent of increasing system state knowledge (both current and predicted future) for all members, and thereby supporting achievement of the team goal. The application of constraints that increase system state knowledge enhances both control and predictability.

The game-play methodology can be applied to all levels of organisation from a team of individual members to a 'team' of Organisational units' that share a common goal. It will become apparent that SBM derives from application of the game-play methodology.

Background of Invention

Known queueing systems of the type referred to above are prone to random customer arrivals (Poisson interarrival pattern) with randomly distributed workloads. System variation makes the allocation of appropriate staffing levels to serve the queue difficult and the dilemma is whether to have less staff operating at high usage rates or more staff at low utilization. The first situation invariably leads to lengthy queues, long customer waits, and reduced customer and staff satisfaction. The second situation involves increased costs, service staff idle for periods of time, staff dissatisfaction but greater customer satisfaction. The trade-off is clear - the cost of increased staff levels, versus the cost of customer dissatisfaction.

Several approaches have been trialled in the OTC service industry - many have been abandoned; some remain but have little impact in addressing the problem.

One prior art proposal involves establishing several queues, each dedicated to a different type of transaction. For example, some banks have set up queues explicitly for deposits and withdrawals while all other banking business is directed to a general queue. The problem remains that not all deposits and withdrawals are quick transactions - random variation in transaction times remains within a single class that is defined by an attribute other than service time. For example coin deposit and foreign currency deposits can take longer than preparing a bank cheque. Further these systems rely on customer decision-making before entering a queue - a reliance that is prone to abuse or mismanagement. A further problem with this approach arises when the number and rates of arrival of the two separate queues are not matched. If no deposit or withdrawal customers arrive in a period in which many 'other type' customers arrive, service operators for the 'quick queue' will remain idle in anticipation of customers, while the 'other type' operators are overloaded.

An extension of this prior art uses historical information on transaction types and service times to determine statistical average service times (and statistical varianceO for particular types of transaction. Service models are then developed to predict service requirements based upon the predicted distribution of transaction types. This approach is fraught with similar uncertainty to that in the original idea of separate queues based on transaction attributes. A statistical average can be a poor representation of a certain attribute - if the statistical standard deviation is large, using the average value for service time will result in poor system performance.

Another approach involves establishing a separate server to deal only with account payments - the intent being to provide a responsive mechanism for these quick transactions. Mostly this can be achieved, but the problem of excessive queueing in the main queue is exacerbated by the reduction in service staff for that queue, which is necessary to establish a dedicated accounts payment server. System variations may also mean that the payments server remains idle in anticipation of the next accounts customer, while the main queue grows.

Lengthy queues still remain in existing queue-server systems. Queueing theory has provided no answer to the problem since its state variable is queue-length - a very poor indicator of system load in that for example, two customers may carry 20 minutes of work, while five customers may carry less than 10 minutes work in total. The length of queue is related to factors external to the queue such as the service rates of the customers currently in the queue and the arrival rates of customers who will join the queue. Queue length provides imprecise and inaccurate information about the stste of the service system.

Summary of Invention

The present invention aims to provide an alternative to known systems and methods for queue-server system management. As used herein the expression "queue" is to be given a broad meaning and is to be understood to refer to situations where customers tasks or services or the like to be performed or carried out sequentially by one or more servers, operators or servicers for a plurality of queuers, ie the customers, who must wait for their respective task or service to be commenced. Thus the word "queue" is not limited to a physical queue in which people stand in a single line or in a number of lines. As used herein the expression "server" is to be given a broad meaning and is to be understood to refer to either an individual or group of individuals who provide a service.

As used herein the expression "queuer" is to be given a broad meaning and is to be understood to refer to a person or organisation or other entity requiring that a task, service or the like be performed, the person/organisation etc being required to wait with others for the service to be performed.

This invention in one aspect resides broadly in a method of managing an over-the-counter queue-server system in which a plurality of queuers wait for a service to be performed by at least one server, the method including:- each operating server ascertaining the nature of a service to be provided, and prior to or soon after commencing the service for a queuer, providing an indicative estimate of the time required for that server to perform the service, the indicative estimate being based on the individual server's knowledge of his or her operating ability and the type of service requested by the queuer, and initiating a service response if the estimate, or if more than one server at least a predetermined number of the estimates, exceeds a predetermined minimum time.

As used herein the expression "service response" is to be given a broad meaning and refers to any action, indication or otherwise which results in a variation being made to an existing service or level of service provided by the servers. In one preferred embodiment the service response increases the number of servers.

It is preferred that the method includes providing feedback to the queuers of the estimate or estimates.

It is also preferred that each server selects one of a plurality of time intervals as the estimate of the time required to complete the service.

It is also preferred that the service response is initiated if the selected time interval, or if more than one server at least a predetermined number of the selected time intervals, is not the shortest of the plurality of time intervals.

It is also preferred that the method including providing feedback to the queuers of the estimated time intervals.

It is also preferred that the estimated time intervals are displayed to the queuers by code means for each time interval.

It is also preferred that there are three time intervals comprising an extended time interval for prolonged service transactions, a medium time interval for lengthy service transactions, and a short time interval for express service transactions.

It is also preferred that the method includes generating an alarm to advise of the need to initiate a service response if the estimate, or if more than one server at least a predetermined number of the estimates, exceeds a predetermined minimum time.

It is also preferred that the method includes updating the estimate provided by the servers if the actual time of serving a queuer exceeds the indicative estimate.

It is also preferred that the method includes generating an alarm to advise of the need to initiate a service response if the selected time interval, or if more than one server at least a predetermined number of the selected time intervals, is not the shortest of the plurality of time intervals. It is also preferred that the method includes evaluating the performance of the queue-server management system.

It is preferred that performance of the queue-server management system is evaluated by determining the actual time taken by a server to perform a service for a queuer and comparing the actual time with the estimate provided by the server. The performance evaluation my also include providing feedback to the server of the comparison.

In another aspect this invention resides broadly in a system for managing an over-the-counter queue-server system in which a plurality of queuers wait for a service to be performed by at least one server, the system including:- estimate indicating means associated with each server whereby each operating server having ascertained the nature of a service to be provided, prior to or soon after commencing a service for a queuer, can provide an indicative estimate of the time required for that server to perform the service, the indicative estimate being based on the individual server's knowledge of his or her operating ability and the type of service requested by the queuer; assessment means to assess whether the estimate, or if more than one server at least a predetermined number of the estimates, exceed a predetermined minimum time, and warning means to warn that a service response should be initiated if the estimate, or if more than one server at least a predetermined number of the estimates, exceeds the predetermined minimum time.

It is preferred that the estimate indicating means include a plurality of respective input means whereby the servers can select one of a plurality of time intervals containing the estimate of the time required to complete the service.

It is also preferred that the assessment means assesses if the selected time interval, or if more than one server at least a predetermined number of the selected time intervals, is not the shortest of the plurality of time intervals.

It is also preferred that the system includes feedback means associated with each server for providing feedback to the queuers of the estimate or estimates.

It is also preferred that the system includes feedback means associated with each server for providing feedback to the queuers of the estimated time interval or intervals.

It is preferred that the feedback means is a coded display to indicate estimated time intervals.

It is preferred that a first code indicates an extended time interval for prolonged service transactions, a second code indicates a medium time interval for lengthy service transactions, and a third code indicates a short time interval for express service transactions. It is also preferred that the system includes performance evaluation means for evaluating the performance of the queue-server management system.

It is preferred that the performance evaluating means determines the actual time taken by a server to perform a service for a queuer, compares the actual time with the estimate provided by the server, and provides feedback to the server of the comparison. Preferably the performance evaluation means includes:- first timing means to determine when a queuer enters a queue; second timing means to determine when a queuer leaves a queue, and third timing means to determine when a service is completed.

The performance evaluation means preferably further includes:- first time interval calculating means for calculating the time interval that a queuer waits in the queue; second time interval calculating means for calculating the actual time taken by a server to perform a service for a queuer, and comparator means for comparing the actual time taken by the server with the estimate provided by the server.

The performance evaluation means may further include analysis means for adaptively updating the performance of the system in light of the evaluation of actual performance. It is preferred that the performance is described as an updated distribution of wait times or a percentage of wait times that are less than a predetermined minimum time that indicates the performance measure for the queue-server system.

A method is disclosed of enhancing the performance of a team of members operating within a system, the method including:- defining a goal to be achieved by the team; analysing the existing team operating system and the team's operation within the system in relation to the goal, the analysis including considering the abilities and capabilities of the team members and the operation of the team members within the system; designing a game-play to achieve the goal, the game-play indicating the constraints to be applied to each team member to increase each member's knowledge of the current and predicted future state of the system and to achieve the goal; redesigning the game-play if necessary to match the abilities of the team members, the system, the operation thereof, or the operation of a team member therewithin, and constraining the team members to operate in accordance with the game-play. The method preferably further includes providing ongoing support for creativity within the game-play environment. The process of design and implementation of a game-play introduces team members to possibilities for operations outside standard practice. Creativity is developed in the design phase so that players are exposed to new individual and team prospects. This process is a captivating one - opportunities for further creative expression emerge from player involvement in the tactics for process improvement and play design, and hence the overall system benefits by providing ongoing support of creativity within a game-play environment.

It is preferred that the goal is identified quantitatively.

Description of Drawings In order that this invention may be more easily understood and put into practical effect, reference will now be made to the accompanying drawings which illustrate a preferred embodiment of the invention, wherein:-

FIG 1 is a diagram of a netball court illustrating the process of a game-play that is described herein; FIG 2 is a schematic diagram illustrating some sequential principles of the game-play methodology;

FIGS 3A, 3B, 3C and 3D are schematic illustrations of the method of the present invention, State Based Management, applied to an OTC banking environment, and

FIG 4 is a schematic block diagram illustrating the system of the present invention applied to the environment of FIGS 3A to 3D.

Description of Preferred Embodiment of Invention At its broadest level the present invention relates to the business/commercial utilisation of what is termed the "Game-play Methodology" - a process improvement methodology for the application of constraints in a human-activity system. The application of constraints in human- activity systems is not a straightforward problem. A broad approach to applying constraints may serve the team well but fails the individuals. A narrow approach that serves the individuals within an organisation, invariably fails at the team level. The game-play methodology provides a balance for both the team and the individuals in the team. Accordingly it will be useful to explain the concept of a game-play.

The game-play methodology has been developed through analysis of the application of constraints in a team sports environment. In team sports such as Netball a game-play is a precise, repeatable sequence of player movements and ball passes to move the ball from a designated position on the court to a scoring position. The constraints applied to the players by the game-play increase the players' knowledge of both the current and predicted future state of play and thereby improve team coordination and effectiveness in achieving the goal (ie scoring). 1 In business a game-play is a set of constraints applied to the members of a team with the intent of increasing system state knowledge (both current and predicted future) for all members, and thereby supporting achievement of the team goal. The designed game-play constraints must recognise the unique abilities and contribution of each member.

Thus a game-play can be regarded as a tactic by which the entire team participates in a constraining, designated play in support of team goals. The constrained, inflexible nature of a game-play enhances current and predicted future internal system state knowledge for all individuals. System state knowledge is used herein to mean 'knowledge of the operating state of the overall system'. In the netball case, a player would have current internal system state knowledge if she knows the state of team play at that instant - where the ball is; who has it; where this is in relation to the goal; where the other players on the team are positioned, what the abilities of the other players are, etc. - knowledge of the state of the entire system, in relation to the goals of the team.

Game-plays must not be confused with game plans. The idea of a game plan is well established in sporting arenas - game plans provide the overall strategy for the sporting team in the application of their skills against the immediate competition. The strategic vision of a game plan is in contrast to the explicit, tactical nature of a game-play.

By way of example, in the sport of Netball a game plan could be as follows:- "The opposition play an aerial, running game with strong, rapid drives into the circle. When we gain possession of the ball, we will slow the play - hold the ball up - and pace the passes to our end of the court. We will not be drawn into playing at their frantic pace."

The above broad instruction of the game plan can be contrasted with the tightly constraining nature of the following game-play. Figure 1 provides a diagrammatic representation of the constraining instructions given below. •On the whistle

-The Goal Defence and Wing Defence lead forward to the transverse line of their scoring third.

-The Wing Attack leads close to the Centre, feints for the pass, circles behind the Centre and moves to the L-H wing area (as back-up).

-The Goal Attack leads into the centre then drives for the R-H wing, as if to take the second pass from the Goal Defence.

•As the Goal Defence approaches the transverse line: -The Centre releases a tight lob to the Goal Defence and drives to the top of the shooting circle. 'When the Goal Defence accepts the pass:-The Goal Shooter drives to the edge of the shooting circle to draw the defence away from the post, turns tightly and leads to the post to accept the long lob from the Goal Defence.

•Notes:-As always, timing is very important in this play. The Goal Shooter must hold until the

Goal Defence has the ball, before initiating her run. -Back-up is provided by the Goal Attack, Center and Wing Attack - so runs must be timed to be clear for the alternate pass, if needed.

Game-play design is a creative process. It looks beyond the range of standard practices, to create a new process approach, and in doing so, establishes within the team a framework for creative design for continuous improvement. Challenging players to think creatively is an intrinsic part of the game-play methodology and the early stages of game-play design lead the innovative change process by example.

In general it can be said that control and predictability are key features of game-plays. By increasing current and predicted future system state knowledge for all members of the team, the game-play methodology attempts to remove internal variation and support the goal directed efforts of the players. As long as the play is goal-directed and appropriate to the skill level of the players, improvement in system performance is assured.

Looking now at the application of game-play methodology to business and commerce it is observed that the game-play methodology is a process improvement methodology for human- activity systems - targeting specifically the application of constraints.

The game-play methodology is an approach to the application of constraints in any human- activity system and as such has considerable commercial potential for business. It is an adaptive process - self-modifying to match the abilities of the individuals and the unalterable constraints of their environment. The core value of a game-play is that it increases internal system state knowledge - both current and predicted future - for all players and the game-play methodology relies on this aspect for its process improvement.

In broad summary as illustrated in FIG 2, the game-play methodology can be described in a preferred embodiment as including the following steps:- 1. Define the team goal. Involve team members in the definition process to encourage creative involvement and ownership of the outcome.

2. Analyse the existing team operating system. Use information and data gathering techniques to describe the extant operating system and evaluate its performance in relation to the team-defined goal.

3. Design the constraints that constitute the game-play. The design criteria for the constraints of a game-play in business are that:

■ The constraints must increase internal (current and predicted future) system state knowledge for all team members. * Every aspect of the play must be designed to support achievement of the goal of the team. A misdirected play is simply constraining not supportive. ^« The design must acknowledge the individual limitations and abilities that of each member of the team and the unalterable constraints of the environment. And,

■ The design activity should allow for the creative involvement of the team members in the game-play design and development process.

4. Examine the requirements of the game-play in relation to the abilities of the players and the unalterable constraints of the working environment, and assign or modify roles accordingly. Redesign the play if necessary to match the abilities of the players. While it is valid to extend a player in the design of a game-play, such extension must be realistically achievable.

5. Constrain the individuals to operate in accordance with the game-play.

Turning now in detail to the application of the above methodology to queue-server system management, it will become evident that the present invention in this preferred aspect thereof, SBM, proposes a process that monitors the current and predicted future states of the service system, and is able to predict the onset of loss of flow in the queue, and provide an appropriate response to avoid excessive customer waits.

Queue 'flow' is a relative state. It describes the rate of exit of customers from the queue. In a production setting one component may leave the buffer each half hour and flow is maintained, while in a banking chamber a customer leaving the queue every 90 seconds may be required for flow to be maintained. In the banking case a reduction in flow occurs when all tellers are working on lengthy or prolonged transactions and are effectively 'blocked' and the queue stops 'flowing' for a period. Flow evaluation is subject to problems of customer perception. A customer who is in a hurry may feel that a queue is 'stagnant' if a customer is exiting each minute, while another customer who is not so pressed for time may be satisfied that a queue is flowing if a customer leaves every two minutes.

The state of the overall system is described by monitoring the individual states of the service providers with categorization of the service states. The categorization process involves consideration of the distribution of service times, the distribution of interarrivals, customer expectations in regard to service (and in particular queue time), and the abilities of service operators to accurately assess and categorize customer workloads. An exemplification of SBM and this categorization process based on a study of OTC transactions in the banking industry now follows.

Based on observations regarding the impact of transaction time on wait time (time in queue), transactions were categorized as illustrated in Table 1. This categorization was developed from analysis of interarrival data [lognormal with variable mean and standard deviation over time] and transaction data [lognormal distribution - mean of 178 seconds; standard deviation of 220 seconds] and the direction given by the defined goal that in this illustrative example is to maximize the number of customers who queued for less than three minutes.

Table 1. An example of categorization of transactions by their service time - OTC banking service.

Having established the categories that will define the state of the individual service operators, a monitoring system is implemented to track the state of the overall system. In the case described above an electronic system to be subsequently described enables the tellers to indicate their current workload, according to the listed categories. When a customer presents work to a teller, the teller assesses its contents and categorizes it. An ET is flagged as green, an LT as orange and a PT as red. The electronic system monitors the overall state of all tellers and displays the individual states to the queued customers. Immediately all tellers are flagged on orange or red, a loss of flow is imminent and a response mechanism is initiated. In this exemplification three states are used to allow for flexibility in the response mechanism. Comparison of only two states (at least one green flag; and no green flags) are used to trigger the response, but the greater detail of three states allows the response mechanism to better assess the current and predicted demands on the system.

The response mechanism can take various forms. In this exemplification the response is to introduce a queue manager and express teller (QMET) team. Another response may be to increase service capacity. Other appropriate responses can readily be determined depending on the circumstances and environment.

The electronic system also monitors the accuracy of the tellers in flagging their workload and updates misflags. Information is also fed back to the tellers to allow them to improve and maintain the accuracy of their workload assessments. Data on transaction and wait times is also collected to allow the demands on the system and the performance of the response strategy to be evaluated.

The invention has allowed staff to improve not only their performance in regard to a team- defined goal: 'maximising the number of over-the-counter (OTC) customers who queue for less than three minutes', but also their efficiency and productivity in related areas.

Improvement can be dramatic - after a four week experimental trial one bank team effected a 143% improvement measured against the goal of maximising the number of customers who queue for less than three minutes. Prior to the introduction of SBM 35% of customers queued for less than three minutes, while 85% enjoyed that level of service after the introduction of SBM. Another trial commenced with higher baseline performance (63%) so that the improvement was less spectacular (87%) but nonetheless significant - 38% improvement within two weeks of application of their game-play.

Reneges at the first mentioned trial dropped from 71 in the pre-game-play week to 4 in the follow-up period, while the maximum waiting time experienced by a customer during the survey period fell from 24 minutes without the play to 12 minutes with the play. (A renege in this context is a customer who abandons current business with the bank. Customers who left the queue to use an alternate channel available on site such as an ATM, telephone banking, flexi-drop box etc were not considered as reneges.)

In the first trial the response mechanism (queue-manager and express teller team) operated for 14% of the observed time, while in the second trial the response mechanism (an additional two tellers - service capacity increased by 50%) operated for an average of 50 minutes per day - less than 13% of the time. In this second trial the three states were used to provide more detailed information to the queued customers although comparison of only two states were needed to trigger the response. There was no reduction in performance of other aspects of the tellers' duties (back-office work and training) as a result of implementing SBM.

In both trials the distribution of service times remained the same indicating that there was no reduction in quality of service as a result of the implementation of SBM.

Turning now to FIGS 3A to 3D which illustrates the practical implementation of the method of the invention in an OTC banking environment, state 1 seen in FIG 3A is the most common operating state - a number of tellers are on the counter and the random distribution of transactions has resulted in at least one teller serving an express customer, at any given time. The team's current knowledge is that teller 2 should take no longer than 90 seconds before she is available to take the next customer. During this time tellers 1 or 3 may also become available, but irrespective of this occurring, the system is still able to maintain 'flow' through teller 2.

State 2 seen in FIG 3B describes the circumstance of all substantive tellers being occupied with either lengthy or prolonged transactions. The instant that the last of the substantive tellers flags on red or orange (i.e. no tellers working on express transactions), a bell is rung to signal system blockage - loss of queue flow. A response mechanism is then initiated. In state 3 illustrated in FIG 3C, a QMET pair is brought on line as the response mechanism in this example. The queue-manager directs the first express customer from the queue, to the express teller, and ascertains the composition of workload in the remainder of the queue. If there are two of more customers with lengthy or prolonged work in the queue, an additional teller is brought on line. Three or more customers with lengthy or prolonged work indicates that additional substantive tellers should be brought on line. The server system is now at full capability in state 4 as seen in FIG 3D, and remains in operation until there are no queued customers.

FIG 4 is a schematic block diagram of the system 10 in accordance with the present invention which manages a queue-server system wherein a queue 11 in which a plurality of queuers 12a, 12b.... 12n wait for a service to be performed by at least one of servers 13a, 13b....13n. Queuers 12 leave queue 11 as shown by arrows and approach counter 19 where they are served by servers 13. They leave the queue as shown when service is completed. System 10 includes estimate indicating means 14a, 14b....14n associated with each server 13a, 13b....13n for the server to provide an estimate 15a, 15b....15n prior to or soon after commencing a service of the time required to complete the service. System 10 also includes assessment means 16 to assess whether an estimate 15, or if more than one server ie servers 13a, 13b....13n at least a predetermined number of estimates 15a, 15b....15n, exceed a predetermined minimum time.

System 10 also includes warning means 17 to warn that the number of servers 13 should be increased if the estimate 15 or estimates 15a, 15b....15n exceed the predetermined minimum time.

System 10 also includes feedback means 18a, 18b....18n associated with each server 13a, 13b....13n for providing feedback to the queuers 12a, 12b....12n of the estimate 15 or estimates 15a, 15b....15n.

The estimate indicating means 14a, 14b....14n are respective input means whereby the servers 13a, 13b....13n can select one of a plurality of time intervals, brackets or categorisations which contain or embrace the estimate 15a, 15b....15n of the time required to complete the service.

Feedback means 18a, 18b....18n are associated with each server 13a, 13b....13n for providing feedback to the queuers 12a, 12b....12n of the estimated time interval categories.

Assessment means 16 assesses if none of estimates 15a, 15b....15n is in the shortest of the time interval categories.

Feedback means 18a, 18b....18n are colour coded displays to indicate the estimated time interval categories by different colours green, orange and red as described above, with a first colour red indicating an extended time interval for prolonged service transactions, a second colour orange indicating a medium time interval for lengthy service transactions, and a third colour green indicating a short time interval for express service transactions.

System 10 also includes performance evaluation means for evaluating the performance of the queue control system. The performance evaluation means comprises a first timing means 20 to determine when a queuer 12a, 12b....12n enters a queue 11 , a second timing means 21 to determine when a queuer 12a, 12b....12n leaves a queue, and a third timing means 22 to determine when a service is completed.

The performance evaluation means also includes time interval calculating means 23 for calculating the time interval that a queuer 12a, 12b....12n waits in the queue and for calculating the actual time taken to by a server 13a, 13b....13n to perform a service for a queuer 12a, 12b....12n. Comparator means 24 then compares the actual time with the estimate 15a, 15b....15n provided by the server 13a, 13b....13n. Feedback of the results of the comparison is then provided to the respective server either electronically, as for example by email or the like, or by means of management advice. It will be appreciated that the actual electronic means by which the invention is performed in its basic principle should be readily apparent to those skilled in the art. Details are not provided here other than to indicate in broad principle that in one preferred embodiment estimate indicating means 14a, 14b....14n are suitable coloured buttons operable by the servers in accordance with their estimate of the time to be taken to complete the service for the queuer who has just presented for service. . Feedback means 18a, 18b....18n is a coloured LED display. Assessment means 16 is suitably a logical circuit or corresponding software for analysing the inputs from estimate indicating means 14a, 14b....14n to determine whether warning means 17, suitable a buzzer, is to be activated. Similarly the performance evaluation can be effected by logical circuitry or suitable software to calculate time intervals from the timing signals provided from timers 20,21,22, i.e. time interval calculating means 23, to identify those time intervals with the particular server serving the particular queuer, and to make the comparison of the actual with the estimated time, i.e. comparator means 24.

Thus it will be appreciated that the present invention in its broadest aspects represents a considerable advance on management of quality within human-activity systems. Currently available process improvement techniques include Quality Circles, Total Quality Management, Total Productive Maintenance, Continuous Process Improvement, Benchmarking, ISO 9000, and Six Sigma.

Most of the current process improvement techniques focus on the form of the process or the product, in regard to product quality. The game-play methodology is a system approach that attempts to consider not just the form of the process and the quality required of the product, but also the contribution of each individual, and the coordination of their efforts.

A number of advantages of the game-play methodology include its broad applicability - any human-activity system can benefit from its application. It is a pragmatic approach to process improvement - avoids abstractions. Its framework is simple to implement in a step-wise process. It addresses the dilemma of applying constraints to the individual so that the overall team effort is supported while sponsoring the individual's freedom to investigate and develop creative responses for the organisation. It promotes goal definition at a tactical level, organises individual involvement in 'process improvement' and 'job design', supports the development of team values, leads to measurable process improvement, extends the abilities of the individual; and supports the creative process.

It will be appreciated that the present invention in its various aspects has a number of advantages over known queue control strategies including the following: . The process has broad applicability

It allows service operator to predict and respond to the onset of blockages in their service system - i.e. loss of queue flow

It is a dynamic, adaptive process that addresses the problem of customer dissatisfaction with excessive waiting for service. . It does not require a reduction in quality of service offered or an increase in staffing levels.

It is simple to implement - training is fundamental and can generally be achieved in less than an hour.

It is operable by both novice and experienced service operators without extra training for the novice. It will of course be realised that whilst the above has been given by way of an illustrative example of this invention, all such and other modifications and variations hereto, as would be apparent to persons skilled in the art, are deemed to fall within the broad scope and ambit of this invention as is herein set forth.

Claims

The claims defining the invention are as follows:-

1. A method of managing an over-the-counter queue-server system in which a plurality of queuers wait for a service to be performed by at least one server, the method including:- each operating server ascertaining the nature of a service to be provided, and prior to or soon after commencing the service for a queuer, providing an indicative estimate of the time required for that server to perform the service, the indicative estimate being based on the individual server's knowledge of his or her operating ability and the type of service requested by the queuer, and initiating a service response if the estimate, or if more than one server at least a predetermined number of the estimates, exceeds a predetermined minimum time.

2. A method of managing a queue-server system as claimed in claim 1 , wherein the service response increases the number of servers.

3. A method of managing a queue-server system as claimed in claim 1 , wherein each server selects one of a plurality of time intervals as the estimate of the time required to complete the service.

4. A method of managing a queue-server system as claimed in claim 3, wherein a service response is initiated if the selected time interval, or if more than one server at least a predetermined number of the selected time intervals, is not the shortest of the plurality of time intervals.

5. A method of managing a queue-server system as claimed in claim 1 , and including providing feedback to the queuers of the estimate or estimates.

6. A method of managing a queue-server system as claimed in claim 3, and including providing feedback to the queuers of the estimated time intervals.

7. A method of managing a queue-server system as claimed in claim 6, wherein the estimated time intervals are displayed to the queuers by code means for each time interval.

8. A method of managing a queue-server system as claimed in claim 7, wherein there are three time intervals comprising an extended time interval for prolonged service transactions, a medium time interval for lengthy service transactions, and a short time interval for express service transactions.

9. A method of managing a queue-server system as claimed in claim 1 , and including generating an alarm to advise of the need to initiate a service response if the estimate, or if more than one server at least a predetermined number of the estimates, exceeds a predetermined minimum time.

10. A method of managing a queue-server system as claimed in claim 1 , and including updating the estimate provided by the servers if the actual time of serving a queuer exceeds the indicative estimate.

11. A method of managing a queue-server system as claimed in claim 4, and including generating an alarm to advise of the need to initiate a service response if the selected time interval, or if more than one server at least a predetermined number of the selected time intervals, is not the shortest of the plurality of time intervals.

12. A method of managing a queue-server system as claimed in claim 1 , and including evaluating the performance of the queue-server management system.

13. A method of managing a queue-server system as claimed in claim 12, wherein evaluating the performance of the queue-server management system includes determining the actual time taken by a server to perform a service for a queuer and comparing the actual time with the estimate provided by the server.

14. A method of managing a queue-server system as claimed in claim 13, and including providing feedback to the server of the comparison.

15. A system for managing an over-the-counter queue-server system in which a plurality of queuers wait for a service to be performed by at least one server, the system including:- estimate indicating means associated with each server whereby each operating server, having ascertained the nature of a service to be provided, prior to or soon after commencing the service for a queuer can provide an indicative estimate of the time required for that server to perform the service, the indicative estimate being based on the individual server's knowledge of his or her operating ability and the type of service requested by the queuer; assessment means to assess whether the estimate, or if more than one server at least a predetermined number of the estimates, exceed a predetermined minimum time, and warning means to warn that a service response should be initiated if the estimate, or if more than one server at least a predetermined number of the estimates, exceeds the predetermined minimum time.

16. A system for managing a queue-server system as claimed in claim 15, wherein the estimate indicating means include a plurality of respective input means whereby the servers can select one of a plurality of time intervals containing the estimate of the time required to complete the service.

17. A system for managing a queue-server system as claimed in claim 16, wherein the assessment means assesses if the selected time interval, or if more than one server at least a predetermined number of the selected time intervals, is not the shortest of the plurality of time intervals.

18. A system for managing a queue-server system as claimed in claim 15, and including feedback means associated with each server for providing feedback to the queuers of the estimate or estimates.

19. A system for managing a queue-server system as claimed in claim 16, and including feedback means associated with each server for providing feedback to the queuers of the estimated time interval or intervals.

20. A system for managing a queue-server system as claimed in claim 19, wherein the feedback means is a coded display to indicate estimated time intervals.

21. A system for managing a queue-server system as claimed in claim 20, wherein a first code indicates an extended time interval for prolonged service transactions, a second code indicates a medium time interval for lengthy service transactions, and a third code indicates a short time interval for express service transactions.

22. A system for managing a queue-server system as claimed in claim 15, and including performance evaluation means for evaluating the performance of the queue-server management system.

23 A system of managing a queue-server system as claimed in claim 22, wherein the performance evaluating means determines the actual time taken by a server to perform a service for a queuer, compares the actual time with the estimate provided by the server, and provides feedback to the server of the comparison.

24. A system for managing a queue-server system as claimed in claim 23, wherein the performance evaluation means includes:- first timing means to determine when a queuer enters a queue; second timing means to determine when a queuer leaves a queue, and third timing means to determine when a service is completed.

25. A system for managing a queue-server system as claimed in claim 24, wherein the performance evaluation means further includes:- first time interval calculating means for calculating the time interval that a queuer waits- in the queue; second time interval calculating means for calculating the actual time taken by a server to perform a service for a queuer, and comparator means for comparing the actual time taken by the server with the estimate provided by the server.

26. A system for managing a queue-server system as claimed in claim 25, wherein the performance evaluation means further includes:- analysis means for adaptively updating the performance of the system in light of the evaluation of actual performance.

27. A method of enhancing the performance of a team of members operating within a system, the method including:- defining a goal to be achieved by the team; analysing the existing team operating system and the team's operation within the system in relation to the goal, the analysis including considering the abilities and capabilities of the team members and the operation of the team members within the system; designing a game-play to achieve the goal, the game-play indicating the constraints to be applied to each team member to increase each member's knowledge of the current and predicted future state of the system and to achieve the goal; redesigning the game-play if necessary to match the abilities of the team members, the system, the operation thereof, or the operation of a team member therewithin, and constraining the team members to operate in accordance with the game-play.