Product traceability is a prerequisite for modern food industry, imposed by legal regulations and commercial pressure from chain partners and society. In principle, however, the availability of traceability does not result in higher revenues. To still get a return on the investment for a traceability infrastructure, the traceability has to be made subservient to the business goals of a company. Consequently, the food industry looks for added value for the inevitable traceability infrastructure. One of the areas where this added value can be found is in improved quality management, of both physiological (intrinsic) product quality (taste, microbiological quality) as of less tangible (extrinsic) product quality (product origin, fair trade). Systematic collection and exploitation of product quality information throughout a chain can support many business functions (including logistics, quality management, process optimisation, and company image). Accurate and non-forgeable measurement of product quality attributes and environment conditions can benefit from developments in rapid method research. This paper introduces the need for detailed quality information in the chain, and describes how rapid methods can be deployed in collecting and processing this information.

Arable farmers and their suppliers, consultants and procurers are increasingly dealing with gathering and processing of large amounts of data. Data sources are related to mandatory and voluntary registration (certification, tracing and tracking, quality control). Besides data collected for registration purposes, decision support systems for strategic, tactical and operational tasks yield enormous amounts of mainly digital information. Data of similar nature but with often varying definitions are collected and processed separately for different purposes. This paper describes for an important arable crop – the processing potato – which data requirements and flows exist at present and how they could possibly be described in a unifying ontology. An ontology describes the concepts, attributes and relations in a specific knowledge domain using a standardized representation language. Important concepts in this domain are for example crop, parcel, soil, treatment and farm. The ontology – once elaborated – will reduce the overlap between information models and helps to overcome the problem of data definition and representation. It is a key element for the development of systems that can automatically learn either with the help of expert knowledge or through adequate numerical techniques.

Studies geven aan dat het onder voorwaarden mogelijk is om privaatrechtelijke initiatieven op het gebied van systematische kwaliteitsbewaking te gebruiken voor een gerichter overheidstoezicht en een beperking van de (administratieve) lasten, ofwel Toezicht op Controle (ToC). De overheid is op zoek naar toezichtsystemen en -arrangementen bij producenten die voldoen aan de gewenste bescherming van de voedselveiligheid, diergezondheid en dierenwelzijn. In dit onderzoek was specifiek de vraag in hoeverre het GMP+-systeem mogelijkheden biedt voor inzet in een ToC-kader

This paper defines a class of problems involving combinations of induction and (cost) optimisation. A framework is presented that systematically describes problems that involve construction of decision trees or rules, optimising accuracy as well as measurement- and misclassification costs. It does not present any new algorithms but shows how this framework can be used to configure greedy algorithms for constructing such trees or rules. The framework covers a number of existing algorithms. Moreover, the framework can also be used to define algorithm configurations with new functionalities, as expressed in their evaluation functions.

In this paper we present a Method for Inductive Cost Optimization (MICO), as an example of induction biased by using background knowledge. The method produces a decision tree that identifies those setpoints that enable the process to produce in as cost-efficient a manner as possible. We report on two examples, one idealised and one real-world. Some problems concerning MICO are reported.

Product traceability is a tool for realising specific goals, such as improved food safety or logisitic optimisation. FoodPrint, a methodological design framework for traceability systems, systematically defines the traceability objectives, and derives design objectives from collective goals of all relevant actors in a chain. During system analysis and design, areas of attention are Process, Information, Organisation and Technology. In a FoodPrint analysis these aspects are integrated in a complete traceability system design. When the design is completed, the following steps, i.e. system development and implementation in practice, can be undertaken.

Het inrichten van een traceerbaarheidssysteem kan uiteindelijk voordelig uitpakken. Traceerbaarheid kan offensief worden ingezet om de eigen bedrijfsdoelen te realiseren, zegt onderzoeker Floor Verdenius van Wageningen UR. De traceerbaarheidsspecialist ziet consumenten niet als hoofddoelgroep van tracking & tracing. Een andere organisatie van de processen in de keten maakt duidelijk waar de winstmogelijkheden liggen