Starburst

'An alternative method for the delivery of mail'

EXECUTIVE SUMMARY

May 2002

Author: Peter Portnoi

Head of Access and Delivery Deployment

The aim of this report is to provide a summary of the trial findings and the recommendations for future Starburst development and deployment.

DEFINITION

Starburst is a delivery method where groups of 2 - 5 staff work together to deliver a number of walks by operating from a vehicle. The group accelerates itself from the office to the start of the delivery in a vehicle that then acts as a nucleus, from which the group operates. The deliveries are broken-down into short 'bursts' that normally start and finish at the vehicle. The 'bursts' take approximately 20 minutes each and this results in staff spending less time working alone and allows a smaller pouch to be used.

BACKGROUND

Starburst 'proof of concept' trials took place two years ago and showed that Starburst had the potential to address the growing problems of weight on delivery and fatigue.

With the prospect of a single, longer delivery span, the TDS team commissioned the latest trials to further test the potential benefits of Starburst.

TRIAL OBJECTIVES

· Test Starburst concept in a broad range of delivery environments.

· Introduce three Starburst delivery groups at each of the seven trial offices.

· Utilise and build upon the methodology used at the proof of concept sites.

· Test whether Starburst is a cost effective alternative.

· Validate the previous findings and produce recommendations that could be used for National deployment.

CONTEXT

The trials were set against the context of mutual recognition that weight on delivery and fatigue was becoming a major issue for both the CWU and Royal Mail.

It was agreed with the CWU to trial Starburst in seven offices, one in each Head of Performance area. These took place between 7th January 2002 and 29th April 2002.

FINDINGS FROM THE TRIALS

The trials were completed successfully because they had unanimous support from the CWU, area management teams and the staff who became part of the Starburst trial groups. This support has resulted in good quality research data being obtained.

The trial finding fall into three broad categories:

People

· Successful Starburst groups are a well matched group of individuals e.g. all full-time staff on five-day weeks.

· In offices where the first group worked well, subsequent groups were introduced smoothly because it sent a positive message to the rest of the office.

· Anecdotal evidence from staff re-confirmed the findings about fatigue.

· Staff felt Starburst also made a significant contribution to personal security and the security of mail.

· Any problems need to receive immediate attention as they are compounded in group working e.g. if one individual is delayed then the rest of the group waits.

· A trained reserve is required to cover absence for performance to be maintained.

Planning

· Starburst is very sensitive to variables in the outdoor environment e.g. housing density, topography.

· Local knowledge is essential during planning and deployment.

· Substantial 'live' post implementation support is required to allow re-configuration.

· It is essential that 'bursts' are equal / balanced because they are sensitive to changes in the number of Special Delivery, Door to Door and packets.

· Starburst reduced the average delivery span.

Vehicles

· Although cost effective Starburst will increase the overall vehicle requirements of the business as it requires a change in the current vehicle profile.

· More drivers will need to be trained because groups need the flexibility of more than one driver.

· Changeover training is required for each new driver at the cost of £200-£260 per individual.

RECOMMENDATIONS

The latest Starburst trials have validated previous findings, it does address weight on delivery, staff fatigue and replenishment costs. However, it does need to be planned effectively because of its sensitivity to factors that can affect its performance.

Many of the issues faced in the trials will be resolved by a full office revision brought about by a changed delivery span.

It is clear that weight on delivery and fatigue will be compounded by a longer delivery span and that Starburst will be a valuable method for tackling these issues and should be used in conjunction with other delivery methods.

Starburst

Final Report

Peter Portnoi

Head of Access & Delivery Deployment

Service Delivery

May 2002

Contents

Page

1 Introduction 3

2 Summary of Recommendations 4

3 Key Issues from the Starburst trial offices

3.1 Planning 6

3.2 Security 7

3.3 Special Delivery 7

3.4 Door to Door 7

3.5 Weight on delivery 8

4 Performance

4.1 Delivery to specification 10

4.2 EP/OP 10

5 People Issues

5.1 IR/ER 10

5.2 General 11

5.3 Contingency Arrangements 12

5.4 Driving 12

6 Equipment 13

7 Vehicles 13

8 Property 14

9 IT Systems

9.1 DPMS 14

9.2 Duty Management (P318s) 15

9.3 DOBI/DODR 15

9.4 Pegasus 15

10 Cost benefit analysis 16

11 Outcome of further work 16

12 Project expenditure 16

13 Conclusion 17

Appendix A - Individual Office Reports and Performance Data 18

1. INTRODUCTION - Starburst

The aim of this report is to provide the sponsors of the Starburst trials with a detailed account of the trial findings and provide recommendations for future development and deployment.

Starburst is a delivery method where groups of 2 - 5 staff work together to deliver a number of walks by operating from a vehicle. The group accelerates itself from the office to the start of the delivery in a vehicle that then acts as a nucleus, from which the group operates. The deliveries are broken-down into short 'bursts' that normally start and finish at the vehicle. The 'bursts' take approximately 20 minutes each and this results in staff spending less time working alone and allows a smaller pouch to be used.

Starburst offers two potential benefits to the business:

· It reduces employee fatigue caused by the increasing weight and volume of mail on delivery.

· It has the potential to offer considerable savings in areas such as acceleration and replenishment during the delivery span.

The diagram provides a simplified model of the Starburst approach.

The trials detailed in this report were undertaken to build on evidence from 'proof of concept' trials completed in several delivery offices which indicated that Starburst had the potential to address the growing problems of weight on delivery and fatigue.

With the prospect of a longer delivery span, the TDS team commissioned these trials to further test the potential benefits of Starburst. It was agreed with the CWU to trial Starburst in seven offices, one in each Head of Performance area. These started W/C 7/1/02 and finished W/C 29/4/02 at the following offices;

· Aylesbury Central DO

· Birch Park DO (York)

· Camberwell DO (London SE 5)

· Chelmsford DO

· Eastbourne DO

· East Kilbride DO

· Weymouth DO

2. SUMMARY OF RECOMMENDATIONS - Starburst

Equipment

· Use of the mini pouch was successful but some standard pouches should be retained to allow for days when bulking out may occur.

· Initially produce prep fillets in monochrome until burst balancing is completed. Final layouts can then be produced in colour.

· Avoid mixing staff on different prep frames or provide prep support to ensure time of completion is uniform throughout the group.

Indoor planning

· During a revision initial routes should be constructed as a traditional delivery and subsequently grouped as Starburst.

· Careful consideration must be given to the first group, as this will be the reference group for the unit.

· During the clear down of the preparation frame, it is recommended that each OPG put the bundles into the colour labelled pouches from their own frame.

· The pouches should be transported to the vehicle, in such a way, to allow loading in reverse sequence into the vehicle, ensuring a logical and efficient work method.

· Special Delivery items, if collected by one group member and then subsequently signed out to the group, will reduce queuing at the Priority Service locker.

· It is important to ensure burst equalisation when delivering Door to Door. Several methods of Door to Door delivery were undertaken at the trial offices, however further evaluation by Industrial Engineers is required to identify the optimum method of delivery.

· Groups of 3 staff proved to be the easiest to construct during the trial.

Outdoor planning

· Initial planning requires a degree of outdoor adjustment. It was most successful where a planner accompanied the group, to facilitate the balancing of the bursts, keeping dead time to a minimum.

· The delivery of large packets and other attendance call items require planning attention to ensure individual bursts are correctly balanced.

Vehicles

· It essential that maximum flexibility of drivers is maintained because each group requires a minimum of two drivers plus floaters and reserves.

· Starburst will increase the requirement of changeover training; currently cessation of changeover training could save the business between £3.5m to £4.5m per annum. (Source Vehicle Services Intranet site). However, a commercial approach such as this needs to be balanced against the health and safety requirements.

· In recognition of the increased demand for 225cf and 400cf crew carriers, and 150cf panel vans. It is recommended that requests for other replacement vehicles be reviewed to better align the fleet. The recommended vehicle types are :-

Group of 2 - 80cf van

Group of 3 - 150cf van or 225 MPV Crew Carrier

Group of 4 - 225 MPV or 400 MPV Crew Carriers.

Group of 5 - 400 MPV Crew Carrier

Information systems

· Duty Management / RCS functionality is required to identify Starburst duties with a unique code.

· DOBI changes will be required to:

· Identify Starburst Groups and Duties

· Identify Crewbuses/ Crew Carriers by type

· Consideration should be given to modifying DODR to identify Starburst group performance rather than the performance of individual walks.

· The delivery route list outlined in the Tools and Techniques Manual replaces the 'Walk Log' for Starburst deliveries.

3. KEY ISSUES FROM STARBURST TRIAL OFFICES

3.1 Planning

Starburst working methodology requires considerable-planning expertise combined with detailed knowledge of the geographical area but this was gained through continuous involvement from the local manager, CWU and staff. Initially, each trial site took one planner five weeks to reach a stage where the first Starburst group could commence. However, this was reduced to approx. 3 weeks when local data was robust and a trained local expert supported the planner.

The significance of involvement was also important in the initial construction of the groups because the manager, CWU and staff provided the detail that resulted in well-matched groups. For example the most successful groups were those who used similar preparation frames and had similar attendance patterns e.g. all full-time and 5 day week duties or all part-time with 6 day week duties.

The trials found that it was important that the initial route structure was built around a traditional method of continuous delivery (when planning a new set of routes under revision conditions). Suitable deliveries for Starburst can then be grouped accordingly and the subsequent bursts constructed. This provides a contingency because it allows a group to be collapsed in an emergency, with one or more of the deliveries being undertaken conventionally.

The trials demonstrated that Starburst methods are very sensitive to variables such as housing density, topography, individual performance, parking restrictions, volume of Special Delivery items, D2D and mail profile. A desktop planning exercise alone could actually increase delivery spans. But this was overcome during the trials by monitoring the Starburst group in the 'live' situation and then re-configuring aspects of the delivery to improve performance.

For example, when one member of a 4 person group was delayed by 2 minutes at each burst, this compounded over the span to over 60 minutes of waiting time (assuming 8 bursts), as the delay of one individual impacted on the entire group. Re-configuration reduced this to a minimum by recording the problems out in the field and subsequently re-balancing. (A walk-balancing sheet is contained within the Tools and Techniques manual).

Excessive packets and small parcels had the same impact, they increased the number of attended calls of one individual so increasing their delivery burst and resulting in a delay for other individuals, within the group. So dedicated packet duties should be carefully reviewed at offices utilising Starburst. Delays will also become an issue if large packet work is introduced to Starburst groups and additionally weight on delivery will increase. Large packets will cause the current Starburst delivery pouch to bulk out so a cost-effective solution needs to be considered.

Key Points

· Considerable planning expertise required.

· Initial make-up of groups is critical.

· Desktop planning is not enough.

· Post implementation adjustments key.

· Review the role of the dedicated motorised packet duties during planning.

· New routes should initially be constructed as a traditional delivery.

3.2 Security

There was a good cross section of delivery areas covered by the trial units, ranging from high income residential areas with limited security risk, to low income high rise time-locked areas where the risk to the safety of staff and mail is high.

In one trial office security escorts were used to ensure the safety of the delivery staff on certain estates because previously staff had been threatened at gunpoint. Following Starburst introduction the staff believe the escorts are no longer required because the security risk has been reduced by splitting the delivery. Each person carries less mail and Starburst ensures that no one is out of sight of another member of the group for more than 20 minutes. Visibility can also be enhanced by the use of High Visibility Vests.

There was some concern about the amount of mail being left in an unattended vehicle. However advice from our Security team shows they were satisfied, that providing correct procedure was used, the vehicle provides an acceptable level of security for normal circumstances.

Key points

· Starburst allows saving to be made where security staff are employed.

· The personal security of staff is improved.

· The security team endorses starburst.

3.3 Special Delivery

The signing for Special Delivery items, at the security locker, was either done by a representative of the Starburst group for all duties or by the individuals concerned, depending on the trial office.

The trial allowed various options, for indoor handling of Special Delivery items, to be evaluated. The conclusion is that there was clear benefit for reducing congestion at the secure locker if one individual collects on behalf of the group and then distributes to the others, using an appropriate signature system.

Special Delivery items were also the cause of down-time at one trial office. Planning took account of the overall volume of Special Delivery items on a delivery or part of a delivery. However, when the items were all concentrated in one burst on one day and in another burst another day, that member of the group was delayed causing the others down-time. Special arrangements may be necessary if this occurs.

Key Points

· A single group member collecting from the locker will reduce congestion but individual accountability for Special Delivery items is essential.

· A process for signing items over from one group member to another should be adopted.

3.4 Door to Door

Door to Door proved to be another area that required careful planning because it created imbalance to the bursts if it is not co-ordinated. Three methods were tried at the trial offices (outlined in the table below). From a Starburst perspective Method 1 was the preferred option because it provided balance and management control but involved some cold calling. Door to Door is an area that requires further investigation by Industrial Engineers before a final recommendation can be made.

	1. One walk per day	2. One row per day per walk	3. Deliver with live mail
+ ·	Bursts equal · Accountability ·	Easy Monitoring	· No cold calling · Bursts equal
-	Cold calling ·	· Cold calling · Uneven bursts · Poor accountability	Poor accountability · Left for floater to deliver.

·

Changes to payment methods were avoided during the trial to minimise any risk of conflict, groups were allowed to make their own decision. However, there is potential benefit when payments are pooled within the group to ensure an even distribution.

Key Point

· Door to Door must be balanced throughout bursts.

· Further evaluation required before determining optimum delivery method.

· Benefits from pooling payments

3.5 Weight on delivery

The Starburst delivery pouch, with the lower weight capacity (11kg), compared to the standard delivery pouch (16kg) greatly reduces employee fatigue on delivery. Average pouch weight per delivery burst was in the region of 8kg. Spot checks gave a range of 6.5kg to 9kg and 11.2kg with D2D.

Comments about feeling less tired were commonplace among staff taking part in the trials. However, fatigue can actually be increased on some deliveries by the removal of cycles and increased walking. Care should therefore be taken to assess suitability.

Although these trials did not seek to evaluate the long-term effects on sick leave, there is evidence from the trial report undertaken by the Human & Environmental Consultancy in July 2000. Which concluded, "The subjects did not report any problems with Starburst. Yet with normal delivery methods, discomfort increases over the span of the delivery." In general the delivery officers commented that they felt less tired with Starburst.

It should be noted that in certain circumstances and/or on particular days of the week the standard size pouch is still required. Bulky but light packets can cause the pouch to bulk out without going overweight and particular problems can be caused by some of the more unusual Door to Door contracts.

The consultants produced the table below, which recommend the maximum pouch weights for current work methods i.e. the ability to carry weight declines as fatigue sets in. This will be a key consideration when increasing delivery spans as it concerns the well being of staff.

Pouch number Recommended weight Cumulative total

1st 16 kg 16 kg

2nd 16 kg 32 kg

3rd 12 kg 44 kg

4th 10 kg 54 kg

5th 10 kg 64 kg

6th 8 kg 72 kg

7th 8 kg 80 kg

8th 8 kg 88 kg

9th 8 kg 96 kg

Pouch number 1st 2nd 3rd 4th 5th 6th 7th 8th 9th

Recommended weight 16 kg 16 kg 12 kg 10 kg 10 kg 8 kg 8 kg 8 kg 8 kg

Cumulative total 16 kg 32 kg 44 kg 54 kg 64 kg 72 kg 80 kg 88 kg 96 kg

Key Points

· In general the lighter pouch reduces fatigue - anecdotal confirms scientific study..

· Under current delivery methodology pouch weight will need to be considered if the delivery span increases.

· Fatigue can actually be increased on some deliveries by the removal of cycles and increased walking so care should therefore be taken to assess suitability.

The following table is the recommended hierarchical approach to deciding the optimum method.

Delivery Method	Reason
High Capacity Trolley Use unless ® ® ® ® ® ® ®	· Terrain is classified as steep (see equipment categorisation table in toolkit)· There are housing types that cannot be serviced within security guidelines (basically can only leave for upto 2 mins in a high risk area).· Pavement width & pedestrian density slow down the walkholder.· There are steps that must be traversed.· Acceleration options result in a more expensive solution than one utilising Starburst.· DP density so low as to warrant cycle use for inter DP travel.
If the HCT unsuitable use the Starburst methodology Unless ® ® ® ® ® ® ® ® ®	· Walks within walking distance of the DO and support can be provided more efficiently than if other methods of delivery used.· Network doesn't support Starburst without resulting in excessive deadtime (for example roads with demand on only one side, networks with lots of vehicle restrictions).· Network doesn't provide sufficient parking spaces due to road restrictions, congestion.· Insufficient walks suitable to make up viable groups· Estate limitations on the number of vehicles that can access the dock.· DP density so low as to warrant cycle use for inter DP travel.
If Starburst unsuitable use the Light Weight Trolley Unless ® ® ® ® ® ® ® ® ®	· Terrain is classified as steep.· There are steps that must be traversed when the trolley will be fully loaded.· DP density so low as to warrant cycle use for inter DP travel.· Replenishment options are available but such a distance from the walk to warrant cycle use.
If the LWT unsuitable then use the High Capacity Cycle Unless ® ® ® ® ® ® ® ® ®	· The terrain is either steep or undulating· DP density is too high, i.e. urban or greater to warrant the use
Use Pouch over the shoulder

4. PERFORMANCE

4.1 Delivery to specification

Performance to at least the existing delivery span was the specification measured at the trial offices. The average delivery span of each individual walk was identified, on a specific day, before it became part of the Starburst group. Then the average span of all the walks in the group was taken on the same day. The Starburst trials aimed to equal or better this average.

Overall, the results were encouraging with three offices showing an improvement and two a marginal improvement. The remaining two offices saw a slight increase in delivery span. However, it should be noted that the improvement in performance was achieved despite constraints, which included no full revision of duties, no dedicated leave reserves and no incentive package.

As with any other revision of duty structures or delivery routes it took a short while for the delivery officer and manager to become fully familiar with the new working practices.

A full report of the trial performance to specification information can be seen in Appendix A

Key Points

· Starburst trials aimed to equal or better the previous delivery span.

· The trials showed encouraging results in performance despite constraints

· Staff and managers will go through a learning curve

4.2 EP/OP

During the trial period members of Operational Research and Analysis and an Industrial Engineer visited three of the sites involved. During these visits they went out on delivery, watched and timed all activities and spoke with many of those involved in the trial. Following the visits, the timings and comments were examined and comparisons made with the planning values within DPMS, which are currently used.

It was concluded that there is no need for a set of planning values for Starburst as the overall effect on measured workload is not significant.

The report's authors can provide the full Operational Research and Analysis "Business Modelling Report on Starburst" if required.

5. PEOPLE ISSUES

5.1 IR/ER

CWU support was extremely positive and the support of the Union at National and local level was invaluable with regard to progressing the trials to timescale. Area management teams were also fully supportive and gave the project teams full co-operation.

Staff perception was variable throughout the trial sites. At some sites there was good-humoured camaraderie at others the persistence of the local manager and CWU representative was required to make Starburst work.

In some case the workload of front-line managers was a barrier because it impacted on the amount of support they gave to the trials. Where the manager was released from some of his day to day work, he or she provided invaluable local knowledge about staff and geography. One manager even went out and drove the vehicle to avoid abandoning a group when there was coverage difficulties.

When the manager's time was more restricted deployment tended to be less successful.

Of particular note are the trials at Camberwell and Aylesbury. The trial was so well received at Camberwell that the intention of the staff and manager is to encompass the Starburst working methodology in at least 50% to 80% of their duties through a revision. Aylesbury have also decided to maximise Starburst throughout the office.

It should be noted that it might not be cost effective for all offices to introduce Starburst to this degree and the table on page 9 Should be used for deciding the optimum method of delivery.

Key Points

· CWU unanimously positive and invaluable to trial.

· Local area management teams supportive.

· Staff perception has been variable.

· The availability of the manager during planning and deployment.

5.2 General

Along with planning, people issues emerged as the key to the success or failure of Starburst at the trail sites. Like any revision of duties, individuals can choose to 'make' or 'not make' a new system work. However with Starburst, if there is a problem, it is compounded because one individual affects the success of the whole group.

It should be noted that under trial conditions, Starburst was imposed on the existing specification and structure. Within a TDS revision, Starburst groups will be constructed specifically to suit the area to be delivered. A number of the problems encountered during the trial would cease to be issues after a revision, as individuals would be picking preferred options. Also a once over the ground delivery structure with a 'job and finish' will offer a considerable incentive which was not available during the trial.

In offices where the first Starburst group worked well, subsequent groups were introduced smoothly. Hence the need to select the groups and geographic areas carefully with full support from the CWU and management. One example from the trial was where a young part-time member of staff completed his delivery by working very quickly and making 20 minutes. When put into a Starburst group he was finishing later because the others were working at a normal pace; understandably he did not want to make the idea work. It is therefore essential that the first group be carefully selected as a reference group to promote Starburst within the unit. It may also be preferable to start with a group of 3 as these proved to give the least initial problems during the trial.

The importance of group selection was highlighted in one case, where an individual in a group 'cut off' for a meal break which meant the entire group returned to the office, delaying the delivery of the final bursts by an hour. Consideration should also be given to staff wanting to work on their own, very slow people who will always hold the group up and those who prefer to take their car on delivery. In these instances it was better to reduce the size of the group rather than fight the individual. Under full deployment and TDS, individuals would be able to select their preferred option as Starburst or non-Starburst so eliminating most of these problems.

There was some dissatisfaction among staff on Saturdays when they traditionally completed their delivery early by going quicker. Starburst does not give individuals this flexibility but this is unlikely to be an issue with TDS if every day is standardised.

Key Points

· TDS will resolve many of the people issues encountered during the trials

· Matching individuals in groups a key to success

· Better to start with groups of three

· Success of first group demonstrates positive message to office

5.3 Contingency Arrangements

The Starburst trials experienced problems with covering absence of group members because a temporary recruit to the group was untrained and unfamiliar with the method of delivery. As a result, three options for coverage are recommended below. However since the trials, the TDS teams have worked up a set of Standard Staffing Modules, which when deployed will overcome any coverage issues.

Option 1

A standard delivery is attached to the group and the duty holder is trained on the Starburst duties. In the event of absence and no suitable reserve available, the standard delivery should be covered by other means and the individual transferred to the Starburst group.

Option 2

One of the walks within a group should be planned as 'removable'. In the event of failure to cover absence by a suitable reserve, this walk can be taken out of the group and covered by other means. The remaining group members will undertake their own bursts as planned and divide the absent petal of each burst between them. With one of the walks now removed the span time of the group should remain broadly similar.

NOTE: It is important that the prep frame of the 'removable' walk can easily be used for a standard delivery method. Also that the number of drivers in the group remains the same.

Option 3

As with option 2 except that all walks remain in the group. Outdoor time will be extended, as group members will now have to cover the absent walk. Overtime could be shared between the group members but it may be necessary to provide prep support to minimise delay to time out.

The Tools and Techniques manual explains coverage in further detail.

5.4 Driving

· There would be a shortfall in drivers if Starburst were fully deployed because it increases the driver requirements with a minimum of two drivers per group plus floaters and reserves. 'Changeover' driver training costs between £200 to £260 per individual, to allow them to drive the 150/225cf vehicles. The same applies when using the 400cf MPV but this size of vehicle also attracts a driving allowance of £23/week per driver so the additional cost to the business needs to be evaluated. 80cf vehicles do not require changeover training, however they only support groups of 2 which will require a greater number of vehicles and bring the associated problems of parking and general vehicle management. Further consultation with health and safety experts will be required regarding 'changeover' training because the commercial needs of the business need to be balanced against health and safety requirements.

Key Points

· Address people issues at outset

· Success of the first group is key to future deployment within the unit.

· Cessation of Changeover training could save the business between £3.5m to £4.5m per annum. (Source Vehicle Services Intranet site). However, a commercial approach needs to be balanced against health and safety requirements.

6. EQUIPMENT

The Starburst trials used a new style of "Mini Pouch" suitable for carrying between 8kg and 11kg on an average burst. Each delivery officer required between eight to ten delivery pouches to carry out their responsibilities within the Starburst group, operating a 150-minute specification.

Existing delivery pouches need to be retained and used for days when an individual Starburst burst 'bulks out' due to bulky packets, D2D items etc.

The fittings were altered temporarily to facilitate the trial, using various methods to mark the frame to identify "bursts" on the walk fillets or using card inserts, sticky tape and plastic inserts. All these 'ad hoc' methods were evaluated during the trials but they would no longer be required following an indoor delivery revision because new colour coded fillets, would be produced from the Minotaur system or the "Prepare and Print Labels" facility within Pegasus.

Under full revision conditions where prep frames have to be changed, it is recommended that the initial set are marked in monochrome until burst balancing is completed and the optimum routing is finalised. The fillets can then be reproduced using the appropriate colour coding to represent the bag colours.

A Microsoft Excel application was developed during the trial for the production of bundle lists, route lists and bag labels. (See associated Tools and Techniques manual for further information).

Delays to 'time out' were experienced when delivery officers completed preparation on different frame types (48 box & RM2000). Staff working on RM2000's waited for colleagues carrying out preparation and bundling on other types of delivery fitting. It is therefore better that the groups include only staff working on similar frames to balance time-out, or adequate provision is made for prep support where justified.

Key Points

· Special pouches have now been purchased.

· Initially produce fillets in monochrome until burst balancing is completed.

· Positive impact on time out if frames changed permanently.

· Better when staff within group is using the same fittings and/or prep balancing is included.

7. VEHICLES

The trials identified that the current national fleet profile of vehicle types will not fully support the expected requirements of Starburst. The norm for Starburst is to have groups of between 3 to 5 staff with a 150cf panel van, suitable for a group of three and a 225cf or 400cf crew carrier, suitable for the groups of four or five. At most offices it was possible to source the right type of vehicle from within the area, by swapping vehicles around and hiring standard replacements. It should be noted that most hire companies do not offer a vehicle suitable for Starburst groups of 4 or 5.

The current fleet profile will dictate the initial planning of group sizes. Group optimisation will be necessary against the overall fleet profile, moving vehicles to different sites to ensure the maximum benefits are achieved.

The use of sub optimal vehicle types was found to limit planning flexibility, required more drivers, increased fleet size and impacted on yard capacity. It is recognised that the fleet profile can only be changed over a period of time, therefore Starburst efficiencies will be reduced as a result.

A breakdown of the Starburst crew carrier at the Eastbourne office highlighted the shortfall of suitable vehicles with no reserve available. It led to two150cf panel vans being used and the need for the office manager, to drive a vehicle to ensure deliveries were made. Under trial conditions this was acceptable but is clearly not feasible in general operation. A vehicle breakdown contingency plan is included within the Tools and Techniques manual.

It is feasible to utilise 80cf vehicles when planning groups of two Starburst deliveries. This approach allows for greater flexibility and makes best use of the current vehicle profile. Planning of routes should include proposals for replacing two 80cf vehicles with one 225cf crewbus, when the two 80cfs come up for replacement.

Key Points

· Vehicle swaps have accommodated trial requirements.

· Investment in future fleet must be modelled.

· Vehicle investment dictates group size.

· Vehicle breakdown contingency plan has been developed

8. PROPERTY

Minor impact on buildings was reported during the trials. Any large-scale implementation of Starburst will increase the impact, for example on the yard / dock capacity because of the number of vehicles. This will have to be assessed on an office by office basis.

Internally, storage of the increased number of delivery pouches ready for loading caused congestion in some offices, indicating that large-scale implementation will have the greatest impact on any space deficient offices.

Starburst worked better when the group members worked on adjacent preparation frames as this aided group working / bundling.

Key Points

· Space requirements of Starburst need to be assessed on an office by office basis during deployment.

9. IT SYSTEMS

Due to the recent FTD revision activity management information was found to be present and accurate and easily accessed on arrival at the trial units.

9.1 DPMS

The OR&S Industrial Engineering report as previously mentioned in section 4.6 concluded the following with regard to Starburst's impact on DPMS.

Overall it is concluded that there is no need for a new set of planning values for Starburst deliveries as the overall effect on measured workload is not significant. The areas that are most affected are within current unmeasured workload and these result in a small reduction in overall workload in DPMS.

The method of setting up DPMS for Starburst is shown on the DPMS Q&A database. By setting up DPMS in this way the overall impact on the DPMS model is not significant enough to impact on issues such as PBS. Indoor workload is not affected at all but outdoor workload will decrease slightly. Only where whole duties can be removed will there be a significant improvement in EP.

9.2 Duty Management (P318s)

Given the impending TDS delivery revision and potential take up of Starburst duties this system will require a total rewrite of existing data. With the underlying reliance on the loshape sub-head codes that underpin both DPMS and RCS, all systems will require an update.

Key Points

· It is recommended that a new RCS code be allocated for "Starburst" deliveries.

9.3 DOBI / DODR

Given the change of fleet profile and equipment and the possible impact on driving duties (reduction in acceleration / replenishment) there needs to be a data update within DOBI. The developers of DOBI may wish to add input screens for "number of Starburst groups" and a break down of "crewbus by type" so allowing any vehicle replacement programme to be co-ordinated.

During the trials it was easy to assess which walks achieved specification but thought needs to be given to the future walk reporting arrangements in DODR. If the delivery specification is to change and methods such as Starburst are to become common then the measures may need to be recorded by Starburst group rather than by the individual duty.

Key Points

· Development of a DOBI screen that will identify the number of Starburst groups and the different Crew Carrier types in the office should be considered.

9. 4 Pegasus

No Pegasus sites were used within trial. However the following recommendations are made based on previously published information:

	IS Planning Toolset	Tabletop	Comments
4 hour walks	As per standard optimisation	As per standard tabletop process	Toolset walks built up from segment info, i.e. length of street, number of DPs, call rate
Segment walks	Run a specific optimisation on 20 minutes	Reliant on existing walkholder and opinion of planners	Toolset will not split codes and therefore some segment >20mins
Plan 'Bursts'	Manual process building on segment information to a maximum of the 4hr span. Repeated for each component walk in turn	Extensive local knowledge for both strands
Update Systems	Labels & PAF for free	Use Aplus & Minotaur as for all walks	Recorded against component walk.

The Toolset approach has significant benefits over Tabletop:

· Objectivity of walks.

· Verified number of delivery points.

· Verified network, including paths.

· Times per street segment.

· Functionality to rapidly run scenarios and maximise the efficiency of the operation.

· PAF benefits, as seamless, rapid, update of PAF without additional keying.

· Label benefits, as the label is automatically created as the walk is planned.

10. COST BENEFIT ANALYSIS

Potential savings may be available from:

· Reduction in replenishment and acceleration duties

· Reduction in security escort duties.

· Cessation of bus fares.

· Reduction in safe drops / pouch boxes / cycles (cost avoidance).

11. OUTCOME OF FURTHER WORK

Trial activities since the release of the interim report include:

· Identifying opportunities to reduce planning time through revised methods

A mentoring / coaching approach can reduce the planning process, this will be utilised at the TDS pilot sites. With support from the Access and Delivery Programme Deployment Team the planning period can be reduced to 3 weeks per group to include initial planning, deployment and re-balancing.(A schematic plan for full deployment is contained within the Tools and Techniques manual).

· Scope the future of preparation frames and Walklogs

The existing APLUS 'Walk Log' system cannot accommodate Starburst. APLUS can only used to update delivery point data in conjunction with prep frame layouts, to maintain PAF and Inward Machine Sorting accuracy. It is recommended that the delivery route list outlined in the Tools and Techniques Manual can now replace the 'Walk Log' for Starburst deliveries.

· Other

· Contingency Arrangements have been included as an independent section within the report.

· A training package has been developed and is now available through the Service Delivery Training Team.

· A Tools and Techniques manual has been developed and a copy is available in conjunction with this report.

12. PROJECT EXPENDITURE

The project has a budget of £140,000

Project spend to 1/5/02 is £100,813

Forecast budget out turn is £100,813

13. CONCLUSION

Starburst has been proven to be a viable and effective method of delivery within certain parameters. It was found to offer a solution to weight on delivery and fatigue but it is important that the criteria around geography, vehicle availability and individual capability are fully explored at the planning stage. Performance improvements are dependent on sufficient time being given to equalising bursts, over the ground, to minimise dead time.

It should also be recognised that the method was vulnerable, as with any other group working activity, to an individual or group of individuals attempting to ensure its failure. Therefore the up front investment of managerial time to achieve the above is essential.