Application of a geometric network solution to the allocation of abattoir samples to regional laboratories.

A large and complex abattoir survey for scrapie in sheep was undertaken in 2002/03 in Great Britain, requiring transport to veterinary laboratories for examination within 3-4 hours. Using a high-resolution road network and data of travel speeds on various classes of roads, travel times between the abattoir and laboratories could be estimated. By incorporating information on maximum laboratory capacity and using a genetic algorithm optimisation routine, an optimum allocation of abattoirs to each laboratory could be developed. Introduction A classic use of geographical information systems (GIS) is for location-allocation problems. These apply when it is necessary to locate a business or service so as to allow maximum access by customers or clients, or alternatively to allocate the latter to fixed services or resources so as to minimise their travel costs. Location-allocation approaches have been applied to assist health planning decisions such as to the optimum siting of hospitals and doctors surgeries and to allocate patients to cancer treatment facilities. Studies of this type have a rich tradition in medical geography, with its concerns for health service resource planning. They also can be considered to fit within the broader discipline of “operational research”, designed to optimise activities such as manufacturing or transport routing. Here we describe the application of these paradigms to a practical veterinary problem, of an abattoir survey aiming to determine the national prevalence of scrapie in sheep. This nationwide survey was initiated in April 2002 and ran until March 2003. It consisted of two components of which that referred to here required the collection of over 20,000 heads from 42 abattoirs. These were then transported to a national network of veterinary laboratories for examination. This transport was subject to considerable logistical constraints. Delivery of heads to the laboratories was required by the afternoon of collection at the abattoir. Contract couriers, priced per mile of transport, provided the transport service and therefore laboratories close to respective abattoirs were preferentially chosen. However, the complex brain sampling protocol, imposed limitations on the potential numbers of heads received by each laboratory. Furthermore, the daily batches of heads from abattoirs could not be divided. Thus, if the capacity of the closest laboratory was exceeded, then a complex reshuffling of other allocations was required. A further logistic problem arose in that the numbers of samples collected and capacity for processing varied from week to week, requiring that decisions of where to allocate samples had to be made at short notice. The project was successfully delivered, on time and within budget. However, the VLA Gisvet Canada AmQ Justifie 9/12/04 2:19 PM Page 33