4.1 The Trialís Foundations
Of fundamental importance in the concept of the trial was the re-use of the existing SaFES off-line product that had been developed by EQMC Ltd., with co-operation from MAFF, and was operational in the Oak Crown Meats plant. The rationale for re-use was two-fold: firstly the off-line system was operational and thus cut down the risk of developing a system from scratch, and secondly the basics of the system had been seen by users and Government bodies and found to meet their needs.
There are several off-line tools on the market for HACCP but they do not extend as far as automatic data monitoring and linking to CCPs. Hardware monitoring, archiving of data and identifying exceptions have been on the market for years but have not been tied directly to a computerised HACCP system as far as we have been able to ascertain.
The way ahead was therefore to base the system on the software module that Oak Crown had been using operationally within their business processes, and thus to concentrate effort on the web-enabled aspects to facilitate remote e-Working with all its advantages. An implication of this was that the intra-modular interfaces were left in tact, in particular EQMC Ltd. were eager that their existing database structure remained and that any new module be bought-in/designed to fit and communicate with that structure. The main emphasis was thus put on the user interface to assess whether the web-enabled tool was a commercially viable product.
From a technical point of view the methodology was as follows: The functionality of the existing SaFES off-line product was taken and a preliminary web-enabled front-end developed. Iterative feedback yielded the front end as seen today. The HACCP Design Tool was implemented and trialled by Oak Crown and Padborg and feedback from Scanfo and EQMC Ltd. The front-end was then integrated with the SaFES system and a single, dual sensored, CCP was monitored at Oak Crown to demonstrate all the links between CCP measurement and archive data retrieval and monitoring.
The Oak Crown system was implemented using a desk top PC based system to aid in the initial development and setting-up processes. It was realised that, even with the off-line system, a monitoring system based around a desk top PC caused operational problems and was seen as at risk from even casual use and tampering. The solution was for the consortium to design a system using a dedicated Single Board Computer (SBC) with input from the data collection modules and output direct to the archived data store.
Thus an SBC system was developed to enable collected data to be directly transferred to the server. Parameter-driven firmware was developed for a commercial SBC supplied by a company specialising in remote data monitoring in the oil industry. The SBC and standard data acquisition units with a full set of sensors were installed at Padborg to allow the users to access the results and monitor whether the system met their requirements.
In this way, we had a PC based system and an SBC at Oak Crown and Padborg respectively thereby allowing us the opportunity to compare the relative merits of the two types of system.
4.2 Software Development
For the Oak Crown system, GUI-generator tools were purchased to eliminate the need for traditional code generation. The methodology employed throughout has been the integration of bought-in modules encompassing hardware, drivers, sensors, GUI-development and web-development tools. The latter two have the advantage of fast GUI generation but the disadvantage of lessened flexibility. So the thrust was on maximisation of existing bolt, customisation and play components. However, concerning the functionality of the SBC system there was nothing on the market that fitted the requirement directly. Consequently, this caused the need to attract a company to develop an SBC system to integrate with the existing database structures. For the Padborg system, many of the SaFES toolís management functions were not required and hence there was no need to invoke the GUI-generated functions embedded in the Oak Crown system.
4.3 e-Working Methods within the Consortium
During the trial we wanted to investigate the impact of the use of electronic interchange as the major mechanism for communications between the partners. The five major partners in the consortium had a physical meeting in Denmark soon after the commencement of the contract (December 2000) and the second full meeting of the consortium was at the Second Project Review at Padborg in May 2002. A further reason was to retain that budget for other activities such as meetings between the design team and suppliers of the GUI generator, Data Acquisition Units and SBC and for review, assessment, dissemination and concertation activities by van Helsing. This approach to distributed working to some extent mirrors the basis of the SaFES on-line tool in that a distributed team can exchange ideas and refine a product by use of e-interchange. This approach was further tested during the installation, commissioning and setting to Work of the system at Padborg. The kit of parts was delivered directly to Padborg after the soak-test activities undertaken by the SBC system supplier in the UK. The installation was thus carried out by the Padborg Poultry Slaughterhouse staff without the need to attend a training course and without anyone from the other consortium partners, or their suppliers, being present. In this way we have ascertained valuable data as to what is needed for any downstream sales and has had the added benefit of Padborg having a far deeper knowledge of their system both at commissioning and for downstream maintenance purposes. The added know-how has thus been increased and accelerated. Training for use of the system by the Emborg and NQA staff was also provided remotely between their normal place of work and van Helsingís office.