CN101299248A - Meat product safety tracing method - Google Patents
Meat product safety tracing method Download PDFInfo
- Publication number
- CN101299248A CN101299248A CNA2008101154868A CN200810115486A CN101299248A CN 101299248 A CN101299248 A CN 101299248A CN A2008101154868 A CNA2008101154868 A CN A2008101154868A CN 200810115486 A CN200810115486 A CN 200810115486A CN 101299248 A CN101299248 A CN 101299248A
- Authority
- CN
- China
- Prior art keywords
- meat product
- node
- information
- quality
- positional information
- Prior art date
- Legal status (The legal status is an assumption and is not a legal conclusion. Google has not performed a legal analysis and makes no representation as to the accuracy of the status listed.)
- Pending
Links
Images
Landscapes
- Management, Administration, Business Operations System, And Electronic Commerce (AREA)
Abstract
The invention discloses a meat product safety retroactive method, including steps of acquiring meat product safety retroactive supply chain scope and hierarchy information; adopting nodes to record the position information of meat product and the state information corresponding to the position, pointing to a novel node if anyone of the locating information of the node and the state information corresponding to the position changes, forming a directed graph denoting the node variation, and the novel node recording the changed position information and the state information corresponding to the position; acquiring the quality safety level of the node according to the position information and the state information in the node; traversing all the nodes of the meat product to be retroacted in the directed graph, locating the node with lowest quality safety level to execute safety trace back. The method provides a method base for the development of meat product safety trace back system which satisfies different retrospective widths and retrospective accuracy requirement, thereby implementing quick locating of meat product supply chain risks and improving trace back and pursuit efficiency.
Description
Technical field
The present invention relates to agricultural system and management engineering technical field, be specifically related to a kind of meat product safety tracing method.
Background technology
Since FAO (Food and Agriculture Organization of the United Nation) in 1974 had proposed " food security " notion for the first time, because the critical incident that food security causes constantly takes place, " food security " become the significant problem that the whole world is paid close attention to.Because agricultural product mostly are the raw materials of food or food, are directly connected to human beings'health and safety, so, also more and more tighter to the technical regulation and the standard of agricultural product both at home and abroad.European Union has formally set up food safety management office of European Union at the beginning of 2002, the overall process of agricultural product from the farmland to the dining table monitored, and require must possess trackability at the agricultural product of European Union member countries' list marketing in legislation in 2005, do not possess import prohibition, the sale of trackability.The U.S. has also issued " food security tracking regulations " in May, 2004, requires all enterprises that relate to food transport, dispensing and import must set up and save from damage the overall process record of related food circulation.China also put into effect agricultural product quality and safeties such as corresponding " agricultural product quality and safety method ", " the departure aquatic products are reviewed rules (trying) ", " herding method " from 2000 can review law.
Under this situation, a large amount of scholars begin can review agricultural product quality and safety and begin one's study, so that follow the tracks of the downstream supply chain of product effectively, recall Related product rapidly, and review the upstream supply chain of product, find the source and the origin cause of formation of problem, to eliminate this problem or to prevent that it from taking place once more.
Existing traceability system mainly is according to reviewing system and legal requiremnt, propose the general frame of traceability system, develop and integrated relevant hardware and software systems, realizing agricultural product from " farmland is to dining table " information integration, and the information inquiry service is provided.T.A.McMeekin for example, technology such as J.Baranyi integrated expert system, RFID, animal identification identification, computing machine have been set up meat quality and can have been reviewed infosystem; The management that is used to trace to the source of Pinto utilization PaniGest software is controlled with quality of assistance.From 2004, China also began to develop successively the agricultural product quality and safety traceability system.At present mainly in large fresh agricultural product fields such as vegetables, aquatic products and meats.Such as: Lu Changhua, Li Xingmin, Xie Jufang utilize different technological development safe pork traceability systems; Li Guangming etc. have inquired into application and the management of RFID technology in food safety trace back; Yin Shouren etc. have studied Beijing's aquatic product quality safety tracing system.
Aspect the modeling and method for solving of the problem of reviewing, according to European Union's definition, can review is the possibility of finding and pursuing quality safety risk factors vestige in supply chain process.From this notion, it is the reticulate texture of a food production to each link of selling that quality safety can be reviewed.How to select appropriate model and derivation algorithm to describe and to review information flow, tap/dip deep into information, so as when problem takes place, to realize fast, accurate location, or the quality safe condition shifted prediction is provided, for quality safety with can review that decision support is provided is crucial problem.Research institution has utilized Gozinto graphs algorithm to improve both at home and abroad, but is limited to the limitation (be limited to three layers and review model) of this algorithm, and main trace back process is not considered links such as market circulation and dispensing, therefore also has certain defective.
Summary of the invention
The purpose of this invention is to provide a kind of meat product safety tracing method, utilize this method to realize the quick location of food supply chain risk, improved the efficient of reviewing and following the trail of, strengthen system to the probabilistic processing power of fuzzy rule, guaranteed the advance of source tracing method and the result's that traces to the source accuracy.
For achieving the above object, the present invention adopts following technical scheme:
The invention provides a kind of meat product safety tracing method, this method may further comprise the steps: defined node, described node are used to write down the positional information and the status information corresponding with this position of meat product; Determine the scope and the hierarchy information of meat product safety tracing supply chain; Scope and hierarchy information based on described meat product safety tracing supply chain make up the digraph that node changes, the variation of the positional information of reaction meat product and the status information corresponding with this position; Reach the quality safety rank of the state information acquisition node corresponding with this position according to the positional information in the described node; Traversal is reviewed meat product each node in described digraph, and the node that the alignment quality level of security is minimum carries out safe tracing.。
Wherein, this method also comprises the optimizing process of simplifying finite graph, this optimizing process can not cause that for the redundant node in the deletion digraph, the positional information and the state information change corresponding with this positional information of described redundant node record the quality of meat product changes.
Wherein, positional information and comprise the different risk factors that cause the meat product mass change in the described node with the corresponding status information of this positional information, other method of quality safety level of being obtained node by described different risk factors is: give different weights according to the extent of injury of described different risk factors, the weighted value that the risk factors that the extent of injury is high are endowed is big, the weighted value that the risk factors that the extent of injury is little are endowed is little, with the summation behind described risk factors and the corresponding weight product, obtain the quality safety rank of described node.
Wherein, described node is the Tuo Ken in the petri net, and the positional information in the node is write down in the transfer in being agree by described holder, and status information corresponding with positional information in the node is write down in the storehouse in being agree by described holder.
Wherein, positional information and the status information corresponding with this positional information in agreeing according to holder are determined the fuzzy prior probability of the meat product quality harm of agreeing in each holder with failure mode effect and hazard analysis method, determine the quality safety rank that holder is willing.
Wherein, in the Petri net, the positional information and the state information change corresponding with this positional information of holding in the palm in agreeing described with state equation, the event information of the generation quality safety harm of holding in the palm in agreeing is described with transition matrix, by described state equation and transition matrix, determine to describe the single step process mass transfer function that meat product quality state transforms, determine the quality safety rank that holder is agree by described single step process mass transfer function.
Wherein, described state equation obtains by meat product current location information and the status information corresponding with this positional information and a last positional information and with the corresponding status information of this positional information, and described transition matrix endangers event information by the generation quality and obtains.
Wherein, determine when the fuzzy prior probability of the meat product quality of Tuo Ken harm with failure mode effect and hazard analysis method, adopt in the reasoning of Bayes's trust network and determine Bayes's trust network node probability based on the algorithm of subduing of heap, obtain the igniting probability that shifts in the Petri net by described Bayes's trust network node probability, obtain the fuzzy prior probability of described meat product quality harm according to described igniting probability.
Utilize food safety trace back method of the present invention, have following beneficial effect: provide the method basis for satisfying the food safety trace back system development that difference reviews width and review accuracy requirement; Realize the quick location of food supply chain risk, improved the efficient of reviewing and following the trail of; Strengthen system to the probabilistic processing power of fuzzy rule, guaranteed the advance of source tracing method and the result's that traces to the source accuracy.
Description of drawings
Fig. 1 is the process flow diagram of meat product safety tracing method of the present invention
Preliminary three grades of classification schemes that Fig. 2 embodiment of the invention forms;
Fig. 3 can review the process flow diagram of unit transition for embodiment of the invention identification;
Fig. 4 is embodiment of the invention node risk confidence level reasoning process figure.
Among the figure: 1, the structure of original decomposition (synthesizing) tree; 2, target is decomposed the structure of (synthesizing) tree; 3, the structure of optimal Decomposition (synthesizing) tree.
Embodiment
The meat product safety tracing method that the present invention proposes is described in detail as follows in conjunction with the accompanying drawings and embodiments.
Be illustrated in figure 1 as the process flow diagram of meat product safety tracing method of the present invention, the method comprising the steps of: S101, defined node, described node are used to write down the positional information and the status information corresponding with this position of meat product; S102 determines the scope and the hierarchy information of meat product safety tracing supply chain; S103 is based on the scope of described meat product safety tracing supply chain and the digraph of hierarchy information structure node variation, the variation of the positional information of reaction meat product and the status information corresponding with this position; S104 is according to the quality safety rank of positional information in the described node and the state information acquisition node corresponding with this position; S105, traversal is reviewed meat product each node in described digraph, and the node that the alignment quality level of security is minimum carries out safe tracing.
This method also comprises the optimizing process of simplifying finite graph, this optimizing process is the redundant node in the deletion digraph, the positional information and the state information change corresponding with positional information of redundant node record can not cause that the quality of meat product changes, when meat product is carried out sectioning, positional information and status information all change, and quality does not change.
Positional information in the node among the present invention and comprise the different risk factors that cause the meat product mass change with the corresponding status information of positional information, as persticide residue, non-ferrous metal content, the time that distance production is experienced etc., other method of quality safety level of being obtained node by different risk factors is: give different weights according to the extent of injury of described different risk factors, the weighted value that the risk factors that the extent of injury is high (promptly causing the high risk factors of harm occurrence frequency) are endowed is big, the weighted value that the risk factors that the extent of injury is little (promptly causing the low risk factors of harm occurrence frequency) are endowed is little, be that weighted value is directly proportional with the harm occurrence frequency, with the summation behind described risk factors and the corresponding weight product, obtain the quality safety rank of described node.
Below in conjunction with description of drawings the specific embodiment of the present invention.
Embodiment
Utilize the characteristics of existing Petri net in the present embodiment, because a holder in the Petri net agree be distributed as two to distribution, can write down the two classes information that is mutually related, therefore the Petri pessimistic concurrency control is introduced in the meat product supply chain, the situation that the harm risk takes place in the dynamic simulation supply chain provides the method basis for satisfying the meat product safety tracing system development that difference reviews width and review accuracy requirement; Utilize failure mode effect and hazard analysis method FMECA then, estimate severity, frequency and the detection degree that supply chain process harm takes place according to the situation that harm risk in the described supply chain takes place, this step provides the fuzzy evaluation foundation for the meat product supply chain emulation based on the Petri net, realized the quick location of meat product supply chain risk, improved the efficient of reviewing and following the trail of, strengthened system the probabilistic processing power of fuzzy rule.
The Petri net that is incorporated in the present embodiment is the mathematical notation to discrete parallel system.The Petri net is to be invented by Ka Er A Petri in generation nineteen sixty, is suitable for describing asynchronous, concurrent computer system model.Because the Petri net can express concurrent incident, be considered to automation theory a kind of (see: woods is rushed. computer network and performance of computer systems evaluation. Beijing: the .2001 of publishing house of Tsing-Hua University; The behavioral theory and the application thereof of the pretty .Petri net of Jiang Chang. Beijing: the .2003 of Higher Education Publishing House; Jiang Zhibin .Petri net and the application in manufacturing system modeling and control thereof. Beijing: the .2004 of China Machine Press; Yuan Chongyi .Petri net principle and application. Beijing: Electronic Industry Press .2005; Wu Zhe brightness .Petri net introduction. Beijing: the .2006 of China Machine Press).
Brief description Petri net can be set up the security traceability network, the scope and the hierarchy information of record meat product safety tracing supply chain, principle and the specific embodiment of the present invention of dynamic simulation meat product supply chain.
With the Petri net write down the meat product safety tracing supply chain scope and hierarchy information, set up the principle that meat product safety can be reviewed network:
The Petri net is a kind of netted information flow model, have storehouse institute and shift two class nodes, in the storehouse and the Tuo Ken (token) that shifts to adding expression quality safety class information on the oriented bipartite graph basis of node distribute, and make driving condition develop, thereby reactive system dynamic running process by certain initiation rule.Given this, adopt the positional information in the transfer record node that holds in the palm in agreeing in the present embodiment, write down status information corresponding in the node by holding in the palm the storehouse in agreeing with positional information, adding holder to according to the quality safety rank of asking the positional information of agreeing in the node to reach the state information acquisition node corresponding with the position agree in the distribution, drive in this enforcement and ask the initiation rule of agreeing change that two classes are arranged, one class is the change of position in the process of circulation, and a class is the change of meat product quality state in the process of circulation.
This specific character of Petri net is widely used in the manufacture field of industry at present, is convenient to the work flow of record product and the additional information of quality.Food processing process in, relate to the record and the state evaluation of machining information equally.The state evaluation here refers to relevant informations such as the work flow of agricultural product in process, quality state.Rely on the node information of Petri net the machining information of agricultural product to be carried out the logic splicing, and agree relevant quality evaluation information that variable can write down agricultural product and output state transition diagram as the holder of record position information, status information by incidence matrix.Above-mentioned incidence matrix is meant the description that the willing position vector of Petri net record holder changes, and by the position transition that the row and column record holder of matrix is agree, is to concern one to one with state transition diagram.By incidence matrix the machining information of agricultural product being carried out the logic splicing is prior art, no longer describes in detail here.Make up the based food quality safety with the Petri net and can review network, this network not only quality state of simulated foods process changes, also comprise emulation to the quality state variation of production, logistics, sales process, therefore, make up the based food quality safety with the Petri net and can review network, not only can follow the tracks of and review, can also carry out record, realize that really the quality of agricultural product can be reviewed the quality state of the agricultural product that are in critical workflow to the machining information of agricultural product.
In the present embodiment at step S102, with the meat product is research object, analyze the meat product quality safety and can review information flow, series relationship information between the business entities such as production, processing, logistics, sale of concrete analysis meat product is determined the scope and the hierarchy information of meat product safety tracing supply chain.
Step S102 is the process of reviewing information gathering and acquisition of information being classified.The information of reviewing is that the information of business entities such as production, processing, logistics, sale is carried out the comprehensive of association link, is that all departments of enterprise coordinate the result that submits to jointly.For authenticity and the completeness that ensures the information of reviewing, requiring enterprise that detailed list is carried out in activity such as the production of agricultural product, processing, logistics, sale writes down or log processing, realize that agricultural product can retrieve the details of upstream or downstream link by data association in different intermediate links, guaranteed agricultural product information in the process of circulation have good grounds, evidence-based.
The present embodiment meat product safety can be reviewed in the information flow three grades of preliminary classifications of formation and review scheme as shown in Figure 2, the first order is for to classify to the meat product that belongs to different enterprises, meat product is divided into A, B, C three classes, is supplied in three different enterprises respectively; The second level is the meat product classification to the inside enterprise that belongs to same enterprise, and corresponding to each different enterprise, meat product can be 1,2 liang of class again in its enterprise; Simultaneously to the further classification of supply of meat product in the enterprise, be divided into 1., 2. two classes.
Like this, the meat product structure of classification and information has gradually just constituted meat product and has reviewed supply chain scope and hierarchy information, certainly meat product is reviewed supply chain and is not limited to three grades of classifications among the present invention, the classification of each grade also is not limited to be divided into two-stage, can classify according to the supply situation of actual meat product, many more supply situations of classifying are detailed more, but it is just complicated more to review supply chain comparatively speaking.
In the present embodiment in step S103, adopt the positional information in the transfer record node that holds in the palm in agreeing, write down status information corresponding in the node by holding in the palm the storehouse in agreeing with positional information, drive in this enforcement and ask the initiation rule of agreeing change that two classes are arranged, one class is the change of position in the process of circulation, one class is the change of meat product quality state in the process of circulation, agree distribution from one of them holder and agree distribution by the next holder of directed arc sensing, like this, utilize the digraph of petri net output just can review supply chain for the dynamic simulation meat product.
Not only can review supply chain for the dynamic simulation meat product with the petri net, can also endanger the situation that risk takes place in the dynamic simulation meat product safety tracing supply chain, locate rapidly and review, specific as follows.
The top priority of reviewing in the present embodiment is, searching causes being reviewed whole risk patterns of meat product qualitative variability, and said here risk pattern is meant and comprises because medicine, environmental pollution, temperature etc. affact the rotten risk factors that change with toxicity of quality of reviewing on the unit to be produced.Here the said unit of reviewing is meant not subdivisible, recordable concrete circulation session information, agree distribute specific to the holder that in the Petri net is record position information and status information.
Utilize petri net to make up the meat product quality safety in the present embodiment and can review network, when endangering the situation of risk generation in the dynamic simulation meat product safety tracing supply chain, various risk factors (variation of the variation of position and meat product quality state) are reduced to storehouse institute and transfer, make the quality safety transitive relation more succinct, be convenient to find the risk circulation way.
Because the Petri net can be set up different system dynamics operational processs according to the different parameters of input in the prior art, the required all kinds of parameters of Petri net are scope and the hierarchy information that obtains the meat product safety tracing supply chain at step S102 in the present embodiment.The meat product safety that the Petri net utilizes the scope of above-mentioned meat product safety tracing supply chain and hierarchy information to set up can be reviewed network model and can be expressed as:
∑＝(P，T，F，W，TL，M
0) (1)
This is a network system of being made up of 6 tuples, symbol " ∑ " library representation institute/transition system, owing to various risk factors in the meat product safety tracing supply chain are reduced in the present embodiment storehouse institute and shift, so claiming the meat product safety of setting up can review network model here is storehouse institute/transition system, P, T, F, W, TL, M
0For constituting storehouse institute/transition system " ∑ " element, data of different types combination in the scope of expression meat product safety tracing supply chain and the hierarchy information is specially:
P={p
1, p
2... p
nBe the finite set of position and the state set that is in each position, describe meat product and review the present intermediateness in unit, wherein p
1, P
2... p
nUse different subscripts to represent that meat product is in different positions, p represents the quality vector of each process, the quality vector here refers to the quality state parameter combination of the description meat product that drawn by the quality evaluation algorithm, has utilized and can realize in the prior art by making up the process that Petri net analog-quality changes.Here said process is meant that meat product reviews the process status of unit, describes the event information that meat product experiences from actual circulation link information aspect;
T={t
1, t
2... t
mBe the limited vector set (status information) of quality transition, and be specially the harm event, be that guiding is exported by incidence matrix by event result, review the information that hazardous events takes place from the food quality aspect, expression quality safety risk produces and situation of change;
F is the set of all directed arcs, represents the information flow direction of meat product, forms the vector of data by the data link of incidence matrix;
W is a weight function, and the weight coefficient of when expression is carried out quantitative test to harm each risk factors being given is if it is higher that the frequency of harm takes place, this weight coefficient is bigger, if it is low that the frequency of harm takes place, this weight coefficient is less, and promptly weight coefficient is to become ratio with the frequency that harm takes place;
TL is record position P for the set of transition time tag
iThe time parameter combination that changes is promptly whenever P
iWhen occurrence positions moved, this set was used for recording time information;
M
0={ M
0(p
1), M
0(p
2) ..., M
0(p
n) be system's initial conditions, expression risk factors distribution situation is wherein distinguished M
0(p
1), M
0(p
2), M
0(p
n) be illustrated respectively in different position p
1, p
2, p
nInitial conditions.
After the information of above-mentioned each parameter is input to existing petri net commonly used, utilize the characteristic that this network can the analogue system dynamic running process, can dynamic simulation meat product safety tracing supply chain in the situation that takes place of harm risk.
Present embodiment is to optimize the meat product quality safety according to the principle that is easy to review most can review network, the Tuo Ken that harm changes does not take place to the process that the Petri net that makes up is optimized for deletion can not cause quality in present embodiment, simplify holder and agree distribute, be specially the flow process that identification can be reviewed the unit transition.
Be illustrated in figure 3 as embodiment of the invention identification and can review the process flow diagram of unit transition, wherein, ground floor A, B, C represent independently enterprise; Enterprises department is represented in the second layer [1], [2], [3], [4].Optimizing preceding traceability system is: the structure 1 that original decomposition (synthesizing) is set is held in the palm willing distribution of a holder of agreeing the distribution sensing C of independent enterprise for one among the A of independent enterprise, 3 willing distributions of 4 holders of holding in the palm among the willing sensing C of independent enterprise that distributes among the B of independent enterprise; Through optimizing process, the structure 2 of target decomposition (synthesizing) tree is transformed to the holder of the willing sensing C of independent enterprise that distributes of a holder among the A of independent enterprise and agree distribution, willing 3 holders of holding in the C of independent enterprise that distribute of 2 holders among the B of independent enterprise agree distribute, be that the holder that a holder among the B of independent enterprise is agree among distribution and the C of independent enterprise agree can not cause mass change, with its deletion; Traceability system after further optimization is optimized at last, the structure 3 that obtains optimal Decomposition (synthesizing) tree agree distribute for one among the A of the independent enterprise holder holder pointing to the C of independent enterprise of agreeing distribute, and 2 holders pointing among the C of independent enterprise of agreeing distribute of 1 holder among the B of independent enterprise agree distribute.By original system change system to the end, the corresponding number of reviewing the unit has tangible minimizing, has improved the accuracy and the validity of system.
With state equation the positional information and the state information change of holding in the palm in agreeing described in the present embodiment, the event information of the generation quality safety harm of holding in the palm in agreeing is described with transition matrix, by state equation and transition matrix, determine to describe the single step process mass transfer function that meat product quality state transforms, determine the quality safety rank that holder is agree by the mass transfer function.
The meat product machining chain is resolved into the combination of several simple process, the quality evaluation information that draws according to the mass transfer function, analyze and to review unit transfer and transition process in every single technology, here said transfer and transition refer to when reviewing the change of target occurrence positions or quality state, promptly take place to shift and the transition process; Based on quality index vector (formula 2), computing mode equation P
nWith transition matrix T
n, determine the mass transfer function.
P
n-1＝{p
n-1，1，p
n-1，2，p
n-1，3，...，p
n-1，k}
P
n＝{p
n，1，p
n，2，p
n，3，...，p
n，m} (2)
T
n＝{t
n，1，t
n，2，t
n，3，...，t
n，i}
Wherein, P
nThe position vector of target is reviewed in expression, wherein, and p
N, 1, p
N, 2, p
N, 3, p
N, mWhat expression was in diverse location reviews the residing status information of target, P
N-1Be P
nBefore continue, the expression P
nThe position vector of laststate, p
N-1,1, p
N-1,2, p
N-1,3, p
N-1, kThe status information that the target laststate is in diverse location is reviewed in expression.T
nThe event information of quality safety risk takes place in expression.P is taking place
N-1To P
nShift movable the time, incidence matrix changes, and then T
nCorresponding incidence matrix changes simultaneously, and incidence matrix is as the input of the evaluation function of petri network internal, draws by Model Calculation and fuzzy evaluation to be in P
nThe safe condition of reviewing target of position, and then can endanger the situation that risk takes place in the emulation meat product safety tracing supply chain.
The situation that takes place according to harm risk in the described meat product safety tracing supply chain in the present embodiment, utilize failure mode effect and hazard analysis method FMECA method to estimate the probability that the harm of meat product safety tracing supply chain process takes place, determine the fuzzy prior probability of meat product quality harm.
Utilize FMECA (i.e. harm pattern influence and harmfulness branch), be a kind of analytical technology with reliability, security, protection (see: Luan Jingen etc. area of computer aided FMECA method. publishing house of Qinhua Univ., Beijing), the limit risk of be used to analyze, auditing system is potential, determine its influence to security of system, thereby find potential weak link in the meat product supply chain, the prevention innovative approach that proposes is to eliminate or to reduce the possibility of harm generation.The principle of FMECA is by the regular record of formulating, the harm pattern that analysis from bottom to top may exist, and the order of severity of harm pattern analyzed, reach a conclusion.Be based on the situation that the harm risk takes place in the meat product safety tracing supply chain in the present embodiment, analyze quality safety risk occurrence cause and result, and carry out explicit description with transfer can reviewing in the network based on the meat product quality safety of Petri net, wherein: hazard ratio is exactly the igniting probability that shifts in the Petri net, the igniting probability promptly may satisfy firing rule and trigger the probability that transition take place, and promptly FMECA provides partly input for the quantitative test of Petri net.
As shown in Figure 4, be example with two factor analyses, node risk confidence level reasoning process extracts three process of meat product processing, represents with A, B, C respectively, considers that single factor 1,2 acts on process A, B, C A
1, B
1, C
1, A
2, B
2, C
3Coordinate system is represented.The x axle is represented technological process, and the y axle is represented the probability that endangers.White portion is represented the safe condition of quality, and the probability of quality harm takes place in the dash area representative.The conversion that acts on the food quality safety of process and harm by factor 1,2 as shown in Figure 3.When carrying out the combined action of influence factor 1 and 2, utilize the FMECA C ' that can produce
1With C '
2(C '
1With C '
2Promptly through the result that food quality is worked the mischief after factor 1,2 actings in conjunction after the 3rd process) acting in conjunction result that food quality is worked the mischief.
In order to determine the probability of reasoning process, select algorithm model commonly used in the reasoning of Bayes's trust network for use: based on the algorithm of subduing of heap, promptly adopted the computing unit of heap, determined trust network node probability by the relation of analyzing between the heap as trust network.Suppose, each node is sought to endanger evidence E, calculating the harm distributed intelligence of asking for node Vq is P (V
q| E), then according to Bayes's posteriority distribution formula
P(V
q|E)＝P(V
q，E)/P(E) (3)
P (V
q, E) be illustrated in node V
qThe probability of generation harm E, P (E) is the ratio of the probability of table harm evidence E generation.Therefore, utilize failure mode effect and hazard analysis method FMECA method to determine the fuzzy prior probability of meat product quality harm.
The fuzzy prior probability of meat product quality harm extends in the Petri network simulation model.(expansion refers in original Petri pessimistic concurrency control, adds the fuzzy prior probability parameter of meat product quality harm) (seeing formula 4) releases basic tracing to the source and tracing algorithm
WFPN＝{P，T，D，I，O，β，α，λ，μ，W} (4)
Wherein: WFPN represents fuzzy prior probability Petri web frame, P={p
1, p
2... p
mBe storehouse institute finite set, be position collection, T={t
1, t
2..., t
nBe the finite set that shifts, for shifting collection, D={D
1, D
2... D
nBe the set of limited proposition, and wherein, proposition D refers to the rule function set of formulating, P, T are consistent with the definition that front Petri nets.I:P → T is an input function, i.e. the mapping that collects from the position collection to transfer; O:T → P is an output function, promptly collects mapping on the collection of position from transfer; β: P → D be from the storehouse to the proposition mapping; α ∈ [0,1] is the degree of confidence according to the definite pairing proposition in storehouse of belief function; λ ∈ [0,1] is the threshold values that distributes according to each transition that the probability function that igniting takes place is determined; μ ∈ [0,1] is the confidence level of the pairing rule of transition; W: to the weight coefficient that shifts T, all input arc weights sums of arbitrary transition equal 1 by position P in expression.
The probability that harm takes place according to the meat product safety tracing supply chain process carries out meat product supply chain risk location, implements reviewing and following the trail of of meat products.
Utilize method of the present invention can realize following function:
Trace to the source model and rudimentary algorithm:
Input: aimed quality point mass vector and all storehouse institute initial mass vector fuzzy probability values; User interactions input fuzzy set;
Output: the quality confidence level vector of each node.
Follow the trail of model and rudimentary algorithm:
Input: the quality vector of certain process node and all storehouse institute initial mass vector fuzzy probability values; The user imports fuzzy set;
Output: the quality confidence level vector of destination node.
In order to verify the practicality of this method in a wider context, this method is used at Shandong Kai Yin group meat production line and Beijing louis d'or quick-frozen food company limited, and is respond well.
Above embodiment only is used to illustrate the present invention; and be not limitation of the present invention; the those of ordinary skill in relevant technologies field; under the situation that does not break away from the spirit and scope of the present invention; can also make various variations and modification; therefore all technical schemes that are equal to also belong to category of the present invention, and scope of patent protection of the present invention should be defined by the claims.
Claims (8)
1, a kind of meat product safety tracing method is characterized in that, this method may further comprise the steps:
Defined node, described node are used to write down the positional information and the status information corresponding with this position of meat product;
Determine the scope and the hierarchy information of meat product safety tracing supply chain;
Scope and hierarchy information based on described meat product safety tracing supply chain make up the digraph that node changes, the variation of the positional information of reaction meat product and the status information corresponding with this position;
Reach the quality safety rank of the state information acquisition node corresponding with this position according to the positional information in the described node;
Traversal is reviewed meat product each node in described digraph, and the node that the alignment quality level of security is minimum carries out safe tracing.
2, meat product safety tracing method as claimed in claim 1, it is characterized in that, this method also comprises the optimizing process of simplifying finite graph, this optimizing process can not cause that for the redundant node in the deletion digraph, the positional information and the state information change corresponding with this positional information of described redundant node record the quality of meat product changes.
3, meat product safety tracing method as claimed in claim 1, it is characterized in that, positional information and comprise the different risk factors that cause the meat product mass change in the described node with the corresponding status information of this positional information, other method of quality safety level of being obtained node by described different risk factors is: give different weights according to the extent of injury of described different risk factors, the weighted value that the risk factors that the extent of injury is high are endowed is big, the weighted value that the risk factors that the extent of injury is little are endowed is little, with the summation behind described risk factors and the corresponding weight product, obtain the quality safety rank of described node.
4, meat product safety tracing method as claimed in claim 1, it is characterized in that, described node is the Tuo Ken in the petri net, utilizes the transfer of described holder in agreeing to write down positional information in the node, utilizes the storehouse of described holder in agreeing to write down status information corresponding with positional information in the node.
5, meat product safety tracing method as claimed in claim 4, it is characterized in that, according to the positional information and the status information corresponding of holding in the palm in agreeing with this positional information, determine to determine the quality safety rank that holder is agree with failure mode effect and hazard analysis method at the fuzzy prior probability of the willing meat product quality harm of each holder.
6, meat product safety tracing method as claimed in claim 4, it is characterized in that, in the Petri net, the positional information and the state information change corresponding with this positional information of holding in the palm in agreeing described with state equation, the event information of the generation quality safety harm of holding in the palm in agreeing is described with transition matrix, by described state equation and transition matrix, determine to describe the single step process mass transfer function that meat product quality state transforms, determine the quality safety rank that holder is agree by described single step process mass transfer function.
7, meat product safety tracing method as claimed in claim 6, it is characterized in that, described state equation obtains by meat product current location information and the status information corresponding with this positional information and a last positional information and with the corresponding status information of this positional information, and described transition matrix endangers event information by the generation quality and obtains.
8, meat product safety tracing method as claimed in claim 5, it is characterized in that, determine when the fuzzy prior probability of the meat product quality of Tuo Ken harm with failure mode effect and hazard analysis method, adopt in the reasoning of Bayes's trust network and determine Bayes's trust network node probability based on the algorithm of subduing of heap, obtain the igniting probability that shifts in the Petri net by described Bayes's trust network node probability, obtain the fuzzy prior probability of described meat product quality harm according to described igniting probability.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CNA2008101154868A CN101299248A (en) | 2008-06-24 | 2008-06-24 | Meat product safety tracing method |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CNA2008101154868A CN101299248A (en) | 2008-06-24 | 2008-06-24 | Meat product safety tracing method |
Publications (1)
| Publication Number | Publication Date |
|---|---|
| CN101299248A true CN101299248A (en) | 2008-11-05 |
Family
ID=40079073
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CNA2008101154868A Pending CN101299248A (en) | 2008-06-24 | 2008-06-24 | Meat product safety tracing method |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN101299248A (en) |
Cited By (9)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN102938091A (en) * | 2012-11-23 | 2013-02-20 | 安徽燕之坊食品有限公司 | Food tracing system capable of ensuring completeness of tracing information |
| CN104376395A (en) * | 2014-09-04 | 2015-02-25 | 太极计算机股份有限公司 | Method and device for tracing quality of iron and steel products |
| CN105677796A (en) * | 2015-12-31 | 2016-06-15 | 山东省标准化研究院 | Food chain network key control node discovery method |
| CN109757401A (en) * | 2019-01-31 | 2019-05-17 | 福建省农业科学院科技干部培训中心 | Based on the method for inhabiting position and individual weight evaluation poultry quality |
| CN109784935A (en) * | 2018-12-04 | 2019-05-21 | 中国农业大学 | A kind of Table Grape supply chain quality safety retroactive method and system |
| CN111274538A (en) * | 2020-01-21 | 2020-06-12 | 中国农业大学 | Method and system for judging quality state of raw aquatic products in storage and transportation processes |
| CN112163866A (en) * | 2020-08-14 | 2021-01-01 | 广州中国科学院软件应用技术研究所 | Market standing book chain construction system |
| CN113474799A (en) * | 2018-12-20 | 2021-10-01 | 艾利丹尼森零售信息服务有限公司 | Food traceability system and method |
| CN115794969A (en) * | 2022-09-08 | 2023-03-14 | 浙江口碑网络技术有限公司 | Distance query method and device and computer equipment |
-
2008
- 2008-06-24 CN CNA2008101154868A patent/CN101299248A/en active Pending
Cited By (14)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN102938091A (en) * | 2012-11-23 | 2013-02-20 | 安徽燕之坊食品有限公司 | Food tracing system capable of ensuring completeness of tracing information |
| CN104376395A (en) * | 2014-09-04 | 2015-02-25 | 太极计算机股份有限公司 | Method and device for tracing quality of iron and steel products |
| CN105677796A (en) * | 2015-12-31 | 2016-06-15 | 山东省标准化研究院 | Food chain network key control node discovery method |
| CN109784935B (en) * | 2018-12-04 | 2021-07-06 | 中国农业大学 | Fresh grape supply chain quality safety tracing method and system |
| CN109784935A (en) * | 2018-12-04 | 2019-05-21 | 中国农业大学 | A kind of Table Grape supply chain quality safety retroactive method and system |
| US11941573B2 (en) | 2018-12-20 | 2024-03-26 | Avery Dennison Retail Information Services Llc | Food traceability system and method |
| CN113474799A (en) * | 2018-12-20 | 2021-10-01 | 艾利丹尼森零售信息服务有限公司 | Food traceability system and method |
| CN109757401A (en) * | 2019-01-31 | 2019-05-17 | 福建省农业科学院科技干部培训中心 | Based on the method for inhabiting position and individual weight evaluation poultry quality |
| CN109757401B (en) * | 2019-01-31 | 2021-08-17 | 福建省农业科学院科技干部培训中心 | Method for evaluating poultry quality based on perching position and individual weight |
| CN111274538B (en) * | 2020-01-21 | 2021-10-26 | 中国农业大学 | Method and system for judging quality state of raw aquatic products in storage and transportation processes |
| CN111274538A (en) * | 2020-01-21 | 2020-06-12 | 中国农业大学 | Method and system for judging quality state of raw aquatic products in storage and transportation processes |
| CN112163866A (en) * | 2020-08-14 | 2021-01-01 | 广州中国科学院软件应用技术研究所 | Market standing book chain construction system |
| CN115794969A (en) * | 2022-09-08 | 2023-03-14 | 浙江口碑网络技术有限公司 | Distance query method and device and computer equipment |
| CN115794969B (en) * | 2022-09-08 | 2023-07-11 | 浙江口碑网络技术有限公司 | Distance query method and device and computer equipment |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| CN101299248A (en) | Meat product safety tracing method | |
| Lima-Junior et al. | Quantitative models for supply chain performance evaluation: A literature review | |
| Sun et al. | Data mining method for listed companies’ financial distress prediction | |
| Narwane et al. | Unlocking adoption challenges of IoT in Indian agricultural and food supply chain | |
| Li et al. | Dealing with uncertainty: A survey of theories and practices | |
| CN106599065A (en) | Food safety online public opinion early warning system based on Storm distributed framework | |
| CN110569904B (en) | Method for constructing machine learning model and computer-readable storage medium | |
| Rožanec et al. | Knowledge graph-based rich and confidentiality preserving Explainable Artificial Intelligence (XAI) | |
| Packianather et al. | Data mining techniques applied to a manufacturing SME | |
| Zhou et al. | Corporate communication network and stock price movements: insights from data mining | |
| Chen | [Retracted] Construction Project Cost Management and Control System Based on Big Data | |
| Guo et al. | A Consumer Behavior Prediction Model Based on Multivariate Real‐Time Sequence Analysis | |
| Montobbio et al. | Labour-saving automation and occupational exposure: a text-similarity measure | |
| Zhang | Modular configuration of service elements based on the improved K‐means algorithm | |
| Kua | On life cycle sustainability unified analysis | |
| US7899776B2 (en) | Explaining changes in measures thru data mining | |
| Rodríguez et al. | Analysis of machine learning integration into supply chain management | |
| Velázquez et al. | Asymmetric competition causes multimodal size distributions in spatially structured populations | |
| Ben Khalifa et al. | Evidential group spammers detection | |
| CN116362371A (en) | Knowledge graph-based purchasing prediction system and method | |
| Zahraee et al. | Applications and challenges of the palm biomass supply chain in Malaysia | |
| Wang et al. | Supply Fraud Forecasting using Decision Tree Algorithm | |
| Tan | Performance evaluation model of cross border e-commerce supply chain based on LMBP feedback neural network | |
| Wang et al. | Research on quality traceability of cigarette by combining PDCA quality cycle with information strategy based on fuzzy classification | |
| Ni et al. | Supply chain demand forecast based on ssa-xgboost model |
Legal Events
| Date | Code | Title | Description |
|---|---|---|---|
| C06 | Publication | ||
| PB01 | Publication | ||
| C10 | Entry into substantive examination | ||
| SE01 | Entry into force of request for substantive examination | ||
| C12 | Rejection of a patent application after its publication | ||
| RJ01 | Rejection of invention patent application after publication |
Open date: 20081105 |