HACCP knowledge modeling and reasoning for agricultural products cold-chain logistics
-
-
Abstract
Abstract: In order to ensure the safety and quality of agricultural products, most agricultural products need to use cold chain logistics in a low temperature environment for transportation, processing and storage. Recently, most studies focus on the safety monitoring and management of a certain single step of cold chain, but a highly effective and reliable cold chain monitoring management needs to realize the knowledge expression, understanding and sharing among multiple agents and business segments in the whole cold chain. The cold chain HACCP (hazard analysis critical control point) knowledge semantic model (CC-HACCP) based on Description Logic SROIQ(D) is proposed to describe the semantic information of the cold chain logistics business and the HACCP management. This model mainly describes the semantic elements such as core concepts, constraints, object attributes, data attributes and attribute characteristics, which are correlated with the domain of cold-chain logistics HACCP management system, and meanwhile, CC-HACCP also uses SWRL rule language to enhance the knowledge reasoning ability. In the experiment, an instance of CC-HACCP ontology was built to describe the business knowledge and rules for the agricultural product cold-chain logistics about "eating oyster", in which the HACCP knowledge sharing requirement between the instances of "step_of_live_oysters_checkup_and_acceptance" and "step_of_live_oyster_transportation" was quite obvious. The model of CC-HACCP was rewritten as an ontology based on semantic web language OWL 2 DL. Through the functions of knowledge checking, case identification and rule reasoning, the HACCP plans in the raw material suppliers, cold chain logistics service providers and production processors were developed, and the new "concept-instance" relationships were automatically identified. For example, "step_of_live_oyster_transportation" was an instance of the concept of cold chain step (CC_Step); after reasoning by reasoning engine, it was also identified as the instance of the critical control point (CCP) and the risky step (Risk_Step). Some new "instance-instance" relationships were also inferred by the HermiT inference engine and the SWRL inference rules. For example, "step_of_live_oyster_transportation" was not required to submit or deliver any proof to the next step, and according to the results of reasoning, the cold chain logistics service provider who was responsible for the "step_of_live_oyster_transportation" would find out that he should generate the proofs of the "transportation monitoring record" and "cargo list", and submit both of them to the next step. Cold chain logistics service provider should also deliver the proofs of "fishing license" and "fishing time identification" from the raw material suppler to production processor. All of the inference processes about above results of the instance of "step_of_live_oyster_transportation" were analyzed in detail in this paper. And according to the transitivity of the property "next_step", all of the subsequent steps of "step_of_live_oyster_transportation" would be found out, and the requires from all of the subsequent steps would be shown in the inference results. These experiment results show that, through the cold chain HACCP management knowledge modeling and the CC-HACCP semantic model, the HACCP plans from different agents can influence each other and be improved, which means that the HACCP knowledge from the different steps in the cold chain has been shared in the whole process. Therefore, the application of HACCP knowledge model has positive significance for the integration and improvement of the multi-HACCP security monitoring and management of agricultural products, which is helpful to improve the efficiency of the cold chain logistics management of agricultural products and ensure the safety and quality of agricultural products.
-
-