Storyboards and User Manual
In trialling the app across different Oxfam regions, we hoped to identify the ways in which the individual shop managers' needs and unique associations could be catered for within such a networked system to realise a more 'human-centric' approach to stock management and scheduling. Shop managers could contribute to the driver optimisation task by using the app to highlight banks for priority collection, even when registering a low fill level due to its geographical location (high yielding stock area) or potential risk of theft if left too long without a service.
They might also stipulate which logistics layer would be most appropriate to make a collection given the dynamic visibility of where the individual logistics layers (local area van driver, shop volunteer) are now, or potentially should be in the immediate future.
Marrying current and historic fill levels of textile/book banks over time and presenting it to users via the app would allow questions like, 'when will the bank be at a sufficient fill level to warrant collection?' This takes the optimisation task to another level where future schedules might be devised based on predicted fill rates and local transport availability. This might see the charity's current logistics layers (area van driver, store volunteer and centralised take-back contractor) used in a more dynamic way, better catering for seasonal changes in donation patterns.
The concept is described pictorially in Figures 1 and 2.
Figure 1. Oxfam stock replenishment scenario
Figure 2. Oxfam stock sharing scenario
The app was designed to have two levels of functionality for two separate user groups: managers and drivers. Drivers acted as the core information gatherers in the system, recording for each bank collection (Figure 3b):
(1) The percentage fill level of the bank on arrival
(2) The number of bags of stock generated (the driver typically empties the contents of the bank and places it into 60-litre hessian sacks to a maximum sack weight of 7.5kg)
(3) The stock quality of the bank (a gauge of the general stock quality on a scale of 1 (poor) to 5 (excellent)
(4) The percentage fill level of the bank at the end of the collection (which might not be zero if there is stock damage or a capacity issue on the van)
(5) Any comments about the collection (this includes text comments and photographs).
At each shop, drivers recorded the numbers of bags transferred as either a delivery of good stock for sale (Figure 3c), sourced from either a bank or another shop, or a collection for 'cascade' to another shop (where the shop concerned had stock, which for a variety of possible reasons had not been sold but could potentially be sold at another location). At the site of a commercial collection or a house clearance, the driver 'added' a new site into the network at that specific geo-location, declaring its name and category, and recording the number of bags/boxes collected by product type along with any comments. To conserve battery life, the app did not continuously track the driver's progress as in a traditional satellite navigation system. Instead, the driver's latitude-longitude location was refreshed whenever the phone was activated in any way, or the driver passed between cell stations of the phone's network provider.
Each time a stock collection was made from a donation bank or from a shop in the form of a cascade, the other members of the community were notified via a push notification from the app to their iPhone or iPad giving the site origin and message 'stock collected available for cascade' (Figure 3d). Push notifications alerted the community of all transactions and messages as they were entered into the system by the respective user. These acted as a temporal reference point for all members of the community, with managers able to view a driver's current position, and then visualise where he or she should be, in hour intervals, during the rest of that day, using a continuously updating geo-location history logged by the system. This feature was designed to allow managers to understand the driver's likely movements, which could be useful, for example, when considering how to respond to incoming house/commercial clearance requests.
The 'heat map' produced (Figure 4a) is a visualisation of the intensity of driver transaction activity (latitude and longitude points where transactions were logged in time) with areas of higher intensity (greater number of visits over time) coloured red, moving through green to blue, indicating lower intensity visitation. With each day's round, more location histories are added to the heat map so that patterns of activity by hour and day can be understood.
A key feature of the app was the messaging system which allowed the members of the community, and assets within it to message each other with requests and notifications. The messaging platform worked on the principle that members of the community represent a specific location or asset and these entities act as the bulletin boards to which messages are attached (Figure 4b). In the case of shops, each manager was registered under their shop and messages were posted to that location in the network. This approach was taken rather than using named individuals because the personnel running the shop and potentially using the app varied from day-to-day and having a general shop bulletin board was considered more flexible. Drivers moving around in the network had messages attached to their map icon by users. Drivers also have the ability to take and post photographs at each bank/shop site which were then added to the message and transaction history of that location (Figure 4c).
As well as being able to view the contents of the driver's van, in terms of the stock held at any point in time, and the transactions as they occurred at banks, shops and other locations, managers could also view the collection/delivery history of each asset in the system. For donation banks equipped with remote monitoring sensors, this allowed managers to receive percentage fill readings twice per day and utilise the information to make better collection scheduling decisions (McLeod et al., 2013). Figure 3a shows that the Sainsbury's clothing donation bank at March was only 28% full on the morning of the 5th June 2013 and did not warrant a collection due to the bank being under 50% full. The app allowed this information to be shared by the members of the community where, in the case of 'shop adopted banks', the shop manager could make a better informed decision as to whether a volunteer needed to visit a bank to prevent it from overflowing.
We understand the extent to which behavioural change in transport habits and practices can be facilitated through the creation of a new form of ‘transport network’, based on extending social networking principles to transport users.
The project has developed a suite of mobile phone apps for each of the corresponding research contexts. Watch videos and read details of the projects aims, key findings and outputs.
The 6ST team comprised researchers from the universities of Southampton, Edinburgh, Salford, Bournemouth and Lancaster.