CN110276112B - Random gradient search optimization method for security inspection system with risk screening mechanism - Google Patents

Random gradient search optimization method for security inspection system with risk screening mechanism Download PDF

Info

Publication number
CN110276112B
CN110276112B CN201910486476.3A CN201910486476A CN110276112B CN 110276112 B CN110276112 B CN 110276112B CN 201910486476 A CN201910486476 A CN 201910486476A CN 110276112 B CN110276112 B CN 110276112B
Authority
CN
China
Prior art keywords
security
risk
passenger
inspection
station
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.)
Expired - Fee Related
Application number
CN201910486476.3A
Other languages
Chinese (zh)
Other versions
CN110276112A (en
Inventor
王嘉宏
吴晓晶
Current Assignee (The listed assignees may be inaccurate. Google has not performed a legal analysis and makes no representation or warranty as to the accuracy of the list.)
Fujian University of Technology
Original Assignee
Fujian University of Technology
Priority date (The priority date 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 date listed.)
Filing date
Publication date
Application filed by Fujian University of Technology filed Critical Fujian University of Technology
Priority to CN201910486476.3A priority Critical patent/CN110276112B/en
Publication of CN110276112A publication Critical patent/CN110276112A/en
Application granted granted Critical
Publication of CN110276112B publication Critical patent/CN110276112B/en
Expired - Fee Related legal-status Critical Current
Anticipated expiration legal-status Critical

Links

Images

Classifications

    • G—PHYSICS
    • G06—COMPUTING; CALCULATING OR COUNTING
    • G06F—ELECTRIC DIGITAL DATA PROCESSING
    • G06F30/00—Computer-aided design [CAD]
    • G06F30/20—Design optimisation, verification or simulation
    • G—PHYSICS
    • G06—COMPUTING; CALCULATING OR COUNTING
    • G06Q—INFORMATION AND COMMUNICATION TECHNOLOGY [ICT] SPECIALLY ADAPTED FOR ADMINISTRATIVE, COMMERCIAL, FINANCIAL, MANAGERIAL OR SUPERVISORY PURPOSES; SYSTEMS OR METHODS SPECIALLY ADAPTED FOR ADMINISTRATIVE, COMMERCIAL, FINANCIAL, MANAGERIAL OR SUPERVISORY PURPOSES, NOT OTHERWISE PROVIDED FOR
    • G06Q10/00—Administration; Management
    • G06Q10/04—Forecasting or optimisation specially adapted for administrative or management purposes, e.g. linear programming or "cutting stock problem"
    • G—PHYSICS
    • G06—COMPUTING; CALCULATING OR COUNTING
    • G06Q—INFORMATION AND COMMUNICATION TECHNOLOGY [ICT] SPECIALLY ADAPTED FOR ADMINISTRATIVE, COMMERCIAL, FINANCIAL, MANAGERIAL OR SUPERVISORY PURPOSES; SYSTEMS OR METHODS SPECIALLY ADAPTED FOR ADMINISTRATIVE, COMMERCIAL, FINANCIAL, MANAGERIAL OR SUPERVISORY PURPOSES, NOT OTHERWISE PROVIDED FOR
    • G06Q50/00—Information and communication technology [ICT] specially adapted for implementation of business processes of specific business sectors, e.g. utilities or tourism
    • G06Q50/10—Services
    • G06Q50/26—Government or public services
    • G06Q50/265—Personal security, identity or safety

Landscapes

  • Engineering & Computer Science (AREA)
  • Business, Economics & Management (AREA)
  • Theoretical Computer Science (AREA)
  • Physics & Mathematics (AREA)
  • Human Resources & Organizations (AREA)
  • Economics (AREA)
  • General Physics & Mathematics (AREA)
  • Tourism & Hospitality (AREA)
  • Strategic Management (AREA)
  • Marketing (AREA)
  • Development Economics (AREA)
  • General Business, Economics & Management (AREA)
  • Geometry (AREA)
  • Game Theory and Decision Science (AREA)
  • Entrepreneurship & Innovation (AREA)
  • General Engineering & Computer Science (AREA)
  • Operations Research (AREA)
  • Quality & Reliability (AREA)
  • Computer Hardware Design (AREA)
  • Evolutionary Computation (AREA)
  • Computer Security & Cryptography (AREA)
  • Educational Administration (AREA)
  • Health & Medical Sciences (AREA)
  • General Health & Medical Sciences (AREA)
  • Primary Health Care (AREA)
  • Management, Administration, Business Operations System, And Electronic Commerce (AREA)
  • Financial Or Insurance-Related Operations Such As Payment And Settlement (AREA)

Abstract

The invention discloses a random gradient search optimization method for a security inspection system with a risk screening mechanism, which comprises the steps of firstly constructing a risk inspection system database and constructing a customs clearance security inspection random restriction type mixed integer type optimization model; then, the optimal risk threshold value tau of the customs clearance security inspection random restricted mixed integer type optimization model is solved by using a random gradient search optimization algorithm*(ii) a Finally, carrying out risk judgment on the passenger to be checked and the carry-on luggage goods by utilizing a random restricted mixed integer type optimization model; if the risk value of the passenger to be checked and the carry-on luggage goods is less than the risk threshold value tau*Making the mobile terminal go to a conventional inspection station with lower inspection capability for security inspection; otherwise, if the risk value of the passenger to be detected and the carry-on luggage cargo is more than or equal to the risk threshold value tau*And going to a strict security check station with stricter check capability for security check. The invention can not only maximize the safety level of the clearance passenger safety inspection system, but also meet the requirement of service efficiency.

Description

Random gradient search optimization method for security inspection system with risk screening mechanism
Technical Field
The invention belongs to the technical field of traffic safety, relates to a random gradient search optimization method for a passenger clearance safety inspection system, and particularly relates to an optimization calculation method for the passenger clearance safety inspection system with a risk screening mechanism.
Background
The security inspection system of the prior art reduces the risk by increasing security inspection equipment or security inspection personnel, however, terrorists are few in the proportion of all customs passengers, so the method inevitably increases a lot of customs security inspection cost and customs clearance time, and the corresponding customs clearance efficiency is reduced. In practice, in the inspection process, a situation of erroneous judgment may also occur, so it is an important technical problem to be solved to ensure the safety inspection level while controlling a certain inspection time and inspection cost.
Disclosure of Invention
In order to solve the technical problem, the invention provides a random gradient search optimization calculation method of a passenger clearance security inspection system with a risk screening mechanism.
The technical scheme adopted by the invention is as follows: a random gradient search optimization method for a security inspection system with a risk screening mechanism is characterized by comprising the following steps:
step 1: constructing a risk checking system database;
step 2: constructing a customs clearance security inspection random restriction type mixed integer type optimization model;
and 3, step 3: solving the optimal risk threshold value tau of the customs clearance security inspection random restricted mixed integer type optimization model by utilizing a random gradient search optimization algorithm*And number of security personnel
Figure BDA0002085553640000011
And 4, step 4: carrying out risk judgment on the passenger to be checked and the carry-on luggage goods by utilizing a risk checking system database;
if the risk value of the passenger to be checked and the carry-on luggage goods is less than the risk threshold value tau*Making the mobile terminal go to a conventional inspection station with lower inspection capability for security inspection; otherwise, if the risk value of the passenger to be checked and the carry-on luggage cargo is more than or equal to the risk threshold tau*And going to a strict security check station with stricter check capability for security check.
Preferably, the customs clearance security inspection random restriction type mixed integer type optimization model in the step 2 is as follows:
maximizing the objective function: d1·R1(τ)+d2·R2(τ)
Satisfies the constraint formula 1: w (τ, s)1,s2)≤ε
The restriction formula 2:
Figure BDA0002085553640000021
restriction formula 3: 0< tau <1
Restricted formula 4: s1,s2∈positive integer
Wherein τ is a risk classification threshold, s, at which clearance passengers are assigned to different security check capability checkpoints1Number of security personnel on duty for conventional security check station, s2For the duty of a strict security check stationChecking the number of people; r1(τ) is the security level of the customs passenger at the conventional security checkpoint when the classification threshold is τ; r2(τ) is the security level of the customs passenger at the strict security checkpoint when the classification threshold is τ; d is a radical of1Accuracy of the security equipment of a conventional security station, d2Accuracy of security equipment of a security station is critical; w () is expected waiting time of the customs passenger in the security inspection system, and epsilon is a required value of efficiency of the expected waiting time of the customs passenger; p (τ) is the distribution probability that a clearance passenger is assigned to a strict security check station when the classification threshold is τ; beta is a1Beta for depreciation and depreciation of security inspection equipment of each year conventional security inspection station2Depreciating and depreciating the cost for security inspection equipment of a strict security inspection station every year; c. C1The on-duty cost of security personnel in a conventional security station, c2The duty cost of security personnel in a strict security station; b is the total budget amount for building a security check station and hiring a security check worker; the positive integer is a positive integer.
Preferably, the specific implementation of step 3 comprises the following sub-steps:
step 3.1: let K be the number of iterations, K the maximum number of iterations,
Figure BDA0002085553640000022
is the upper bound of the threshold value of the risk classification gate, tau is the lower bound of the threshold value of the risk classification gate, M is the number of repetitions of the simulation test, theta1And theta2Improving the gradient parameters of the solution for each iteration;
step 3.2: according to the total budget limit of customs
Figure BDA0002085553640000023
Selecting s1And s2Set of all possible combinations of
Figure BDA0002085553640000024
Step 3.3: setting the iteration number n of the optimization solution to be 0, and starting to execute the following iteration;
step 3.3.1: from the collection
Figure BDA0002085553640000025
Get a set of solutions
Figure BDA0002085553640000026
The number of algorithm iterations k is set to 0, and τ is set1＝τ0；
Step 3.3.2: set k ← k +1 and
Figure BDA0002085553640000027
according to τkPerforming M repeated computer simulation tests, and recording the obtained clearance time
Figure BDA0002085553640000028
And calculating the average value thereof
Figure BDA0002085553640000029
And
Figure BDA00020855536400000210
95% confidence interval of
Figure BDA0002085553640000031
Wherein γ is the half-length of the confidence interval;
step 3.3.3:
if ε satisfies the inequality
Figure BDA0002085553640000032
Or when the number of algorithm iterations K equals K, then set
Figure BDA0002085553640000033
And update the set
Figure BDA0002085553640000034
Otherwise, it is set
Figure BDA0002085553640000035
Wherein theta is1When the gate valve value epsilon of the expected closing time limiting type is larger than the upper limit value of the confidence interval, the gradient parameter is used; and theta2When the threshold value epsilon is smaller than the upper limit value of the confidence interval, the gradient parameter is used; and rotating to execute the step 3.3.2;
step 3.3.4: when the collection
Figure BDA0002085553640000036
Then, the loop of the main program of the gradient search algorithm is terminated and step 4 is entered; otherwise, setting an iteration number n ← n +1 of the optimized solution, and performing step 3.3.1 in a rotating manner;
step 3.4: for the solution obtained by the gradient search algorithm
Figure BDA0002085553640000037
Verifying the feasibility of the program judgment solution by using the feasibility in the sequencing and selecting program, and deleting the infeasible solution under the statistical guarantee; order to
Figure BDA0002085553640000038
All selected by the feasibility verification program are shown
Figure BDA0002085553640000039
Forming a solution set;
step 3.5: obtaining an optimal risk gate valve value
Figure BDA00020855536400000310
And the optimal safety level of the customs clearance security inspection system is SL (tau)*)。
The invention has the beneficial effects that:
1. the random gradient search optimization method provided by the invention is characterized in that the complexity optimization problem existing in a random system is analyzed in a mode of constructing a computer simulation model of a customs passenger security inspection system with a risk screening mechanism, and a customs passenger security inspection random restricted mixed integer optimization model is solved from the computer simulation models of various customs passenger security inspection system design schemes by utilizing a random gradient search optimization algorithm of the patent under the support of a random simulation sample sampling method and a simulation optimization theory, so that the customs passenger security inspection system design scheme which can enable the system service performance expected value to reach the optimum is found out and is provided for a decision maker to make reference.
2. The simulation optimization method provided by the invention can analyze the complex problem that the general mathematical model can not be analyzed, and can analyze the evaluation index and the confidence level of the clearance passenger safety inspection system through the computer simulation experiment design.
3. The invention develops an efficient random gradient search optimization algorithm to solve the problem of the optimized clearance passenger security inspection system with randomness, and can write the simulation optimization algorithm provided by the invention into a software program, and carry out scientific calculation and system simulation through a computer to efficiently solve the random optimization problem.
Drawings
FIG. 1 is a schematic diagram of an embodiment of the present invention.
Detailed Description
In order to facilitate understanding and implementation of the present invention for persons of ordinary skill in the art, the present invention is further described in detail with reference to the drawings and examples, it is to be understood that the implementation examples described herein are only for illustration and explanation of the present invention and are not to be construed as limiting the present invention.
The safety level of the passenger clearance safety check system is maximized under the condition of meeting the allowable clearance time and the customs total budget limit. Under the condition of meeting the requirement of clearance efficiency, the safety level of the safety inspection system is considered to be maximized, and the risk value is used as the basis for assigning passengers to be inspected (and carry-on luggage goods) to various safety inspection stations, passengers with high risk (and carry-on luggage goods) are assigned to carry out strict inspection measures to reduce the risk, and passengers with low risk (and carry-on luggage goods) are subjected to simplified inspection procedures to pass through the safety inspection system quickly, so that good risk management of the safety inspection system is achieved. The technical scheme of the invention considers two decision variables of the optimization model, one is a discrete variable, and the other is a continuous variable, which can be regarded as the optimization problem of the mixed integer decision variable. How to efficiently solve such an optimization problem with mixed integer decision variables remains a difficult technical problem.
The passenger clearance security check system considered by the technical scheme of the invention comprises two check stations with security check capability, wherein the check station with the first security check capability (namely a 'conventional check station') can check passengers (and carry-on luggage goods) more loosely, and the check station with the second security check capability (namely a 'strict security check station') can check more strictly. In the clearance safety inspection system optimization model of the technical scheme of the invention, if the risk value of the passenger to be inspected (and the carry-on luggage cargo) is smaller than the risk screening mechanism threshold value tau, the passenger goes to a conventional inspection station with lower inspection capability for safety inspection; on the contrary, if the risk value of the passenger to be checked (and the carry-on baggage) is greater than the risk screening mechanism gate threshold τ, the passenger goes to a strict security check station with strict check capability for security check. Thus, the passenger under inspection for clearance (with their carry-on baggage cargo) will have a probability p (τ) assigned to a "strict security checkpoint" where security is more stringent, while a probability of 1-p (τ) will be assigned to a "conventional checkpoint" where security is relatively relaxed. Customs will assign the passenger to be checked (and the baggage cargo) to the inspection station with corresponding security inspection capability according to the pre-checked risk value, and the inspection time of customs will also be affected by the number of security inspectors and the number of security inspection devices at the inspection station.
It is desirable to maximize the security level of the entire security check system by adjusting this threshold value while meeting the latency constraint. Through the risk checking information system database, the value is assigned in advance according to the class characteristics of the passenger to be checked (and the carry-on luggage cargo), and then the passenger to be checked is distributed to the checking stations with different security checking capabilities, so that the security level of the security checking system can be maximized under the condition of meeting the allowable waiting time and the total budget limit. If the clearance passenger can be properly assigned to the inspection station with proper security inspection capability, the safety level of the clearance passenger safety inspection system can be maximized, and the requirement of service efficiency can be met.
The random gradient search optimization algorithm provided by the invention is a simulation optimization algorithm for solving a customs clearance security inspection optimization model, and the gradient search algorithm is applied to each simulationThe iteration will continue to search the risk threshold τ until the average pass time W (τ, s)1,s2) Within the confidence interval.
Referring to fig. 1, the random gradient search optimization method for a security inspection system with a risk screening mechanism of the present invention includes the following steps:
step 1: constructing a risk checking system database;
when a passport or an identification document is presented by a customs traveler, personal data is transmitted to a CAPPS auditing System database, and customs personnel assign a risk value corresponding to each customs traveler according to attributes such as the identity background, the existence of a criminal prior subject, historical data and the like of the customs traveler and use the risk value as a decision reference value of a subsequent customs security inspection System.
And 2, step: constructing a customs clearance security inspection random restriction type mixed integer type optimization model;
in this embodiment, the customs clearance security inspection random restriction type mixed integer type optimization model is as follows:
maximizing the objective function: d1·R1(τ)+d2·R2(τ)
Satisfies the constraint formula 1: w (τ, s)1,s2)≤ε
The restriction formula 2:
Figure BDA0002085553640000051
restriction formula 3: 0< tau <1
Restricted formula 4: s1,s2∈positive integer
Wherein τ is a risk classification threshold, s, at which clearance passengers are assigned to different security check capability checkpoints1Number of security personnel on duty for conventional security check station, s2The number of security check personnel on duty of the strict security check station; r1(τ) is the security level of the customs passenger at the conventional security checkpoint when the classification threshold is τ; r2(τ) is the security level of the customs passenger at the strict security checkpoint when the classification threshold is τ; d1Is a general medicineAccuracy of security equipment of inspection station, d2Accuracy of security equipment of a security station is critical; w () is expected waiting time of the customs passenger in the security inspection system, and epsilon is a required value of efficiency of the expected waiting time of the customs passenger; p (τ) is the distribution probability that a clearance passenger is assigned to a strict security check station when the classification threshold is τ; beta is a1Depreciation and depreciation cost beta for security inspection equipment of each year conventional security inspection station2Depreciating and depreciating the cost for security inspection equipment of a strict security inspection station every year; c. C1The duty cost of the security personnel in the conventional security inspection station, c2The duty cost of security personnel in a strict security station; b is the total budget amount for building a security check station and hiring a security check worker; the positive integer is a positive integer.
The modeling concept of this mathematical model is described as follows:
1. the objective function of the optimization model is to maximize the overall security level of the clearance passenger security inspection system: obtaining the distribution proportion R of passengers through a risk screening mechanism door threshold tau1(τ) and R2(τ)。
2. The first limit formula guarantees the requirement of security check time: mean waiting time W (tau, s) for passing passengers1,s2) Less than a tolerable security check latency value epsilon.
3. The second constraint is the total budget constraint: the sum of the costs of the security personnel servicing the equipment, the configuration of the security equipment and the costs of operating the security stations must not exceed a given total budget B.
4. And (3) the third formula limits the assignment range of the risk classification threshold value tau: the risk classification threshold τ is adjusted to be between 0% and 100%.
5. The fourth restriction expression indicates that the number of security check personnel and the number of security check equipment configured for the two types of security check stations are positive integers.
6. The mathematical optimization model takes the maximization of the overall safety level of the passenger clearance security inspection system as an optimization target, and one of decision variables is a gate valve value tau of a customer to be classified, which is assigned to different service type service equipment and belongs to a continuous variable; and the other decision variable is the configuration of the service personnel and service equipment of the two service typesNumber s1And s2Belong to discrete variables.
In this mixed integer optimization model with random restriction, the mathematical function R1(τ),R2(τ), p (τ) and W (τ, s)1,s2) When the risk classification threshold value can not be expressed in a mathematical analytic expression, the invention provides a simulation optimization algorithm, finds out a mathematical function estimated value through a system simulation method, and calculates the risk classification threshold value tau and the personnel allocation number s of two security check stations1And s2The optimal solution of (1).
Even sometimes in some special cases, the function R1(τ),R2(τ), p (τ) and W (τ, s)1,s2) The function value can be calculated in a mathematical analysis mode, but the solving time of the optimization model is greatly increased because the function is too complex and the numerical value is difficult to calculate by a computer. The problem of random optimization is that a function value cannot be expressed by a mathematical expression, and the problem can only be solved by using a simulation optimization algorithm provided by the invention.
And step 3: solving the optimal risk threshold value tau of the customs clearance security inspection random restricted mixed integer type optimization model by utilizing a random gradient search optimization algorithm*And number of security personnel
Figure BDA0002085553640000061
The specific implementation comprises the following substeps:
step 3.1: let K be the iteration number, K be the maximum iteration number, τ be the upper bound of the risk classification gate threshold, τ be the lower bound of the risk classification gate threshold, M be the number of iterations of the simulation test, θ1And theta2Improving the gradient parameters of the solution for each iteration;
step 3.2: according to the total budget limit of customs
Figure BDA0002085553640000071
Selecting s1And s2Set of all possible combinations of
Figure BDA0002085553640000072
Step 3.3: setting the iteration number n of the optimization solution to be 0, and starting to execute the following iteration;
step 3.3.1: from the collection
Figure BDA0002085553640000073
Get a set of solutions
Figure BDA0002085553640000074
The number of algorithm iterations k is set to 0, and τ is set1＝τ0；
Step 3.3.2: set k ← k +1 and
Figure BDA0002085553640000075
according to τkPerforming M repeated computer simulation tests, and recording the obtained clearance time
Figure BDA0002085553640000076
And calculating the average value thereof
Figure BDA0002085553640000077
And
Figure BDA0002085553640000078
95% confidence interval of
Figure BDA0002085553640000079
Wherein γ is the half-length of the confidence interval;
step 3.3.3:
if ε satisfies the inequality
Figure BDA00020855536400000710
Or when the number of algorithm iterations K equals K, then set
Figure BDA00020855536400000711
And update the set
Figure BDA00020855536400000712
Otherwise, it is set
Figure BDA00020855536400000713
Wherein theta is1When the gate valve value epsilon of the expected closing time limit type is larger than the upper limit value of the confidence interval, the used gradient parameter is used; and theta2When the threshold value epsilon is smaller than the upper limit value of the confidence interval, the gradient parameter is used; and rotating to execute the step 3.3.2;
step 3.3.4: when the collection
Figure BDA00020855536400000714
Then, the loop of the main program of the gradient search algorithm is terminated and step 4 is entered; otherwise, setting an iteration number n ← n +1 of the optimized solution, and performing step 3.3.1 in a rotating manner;
step 3.4: for the solution obtained by the gradient search algorithm
Figure BDA00020855536400000715
Verifying the feasibility of the program judgment solution by using the feasibility in the sequencing and selecting program, and deleting the infeasible solution under the statistical guarantee; order to
Figure BDA00020855536400000716
All selected by the feasibility verification program are shown
Figure BDA00020855536400000717
Forming a solution set;
step 3.5: obtaining an optimal risk gate valve value
Figure BDA00020855536400000718
And the optimal safety level of the customs clearance security inspection system is SL (tau)*)。
And 4, step 4: carrying out risk judgment on the passenger to be checked and the carry-on luggage goods by utilizing a risk checking system database;
by using a Computer-Assisted Passenger screening System (CAPPS) in the prior art, when a passport or an identification document is presented by a customs traveler, personal data is transmitted to a CAPPS screening System database, and customs personnel assign a risk value corresponding to each customs traveler according to attributes such as the identity background of the customs traveler, the existence of a pending prior art, historical data and the like to serve as a decision reference value of a subsequent customs security screening System.
If the risk value of the passenger to be checked and the carry-on luggage goods is less than the risk threshold value tau*Making the mobile terminal go to a conventional inspection station with lower inspection capability for security inspection; otherwise, if the risk value of the passenger to be detected and the carry-on luggage cargo is more than or equal to the risk threshold value tau*And going to a strict security check station with stricter check capability for security check.
The technology of the invention focuses on improving the risk management of the clearance passenger security check system, namely classifying clearance passengers (and carry-on luggage cargos) with different risks to a proper security check station, rather than reducing the risk of the security check system by blindly increasing the complexity of security check or increasing security check equipment and security check manpower by consuming huge amounts. In processing an optimization problem with discrete variables, we need to estimate a sublevel function (subvariant), and the prior art often uses a Finite difference method (Finite Differences), which often costs a great deal of computer simulation time. The invention mainly uses the optimization technique of random gradient search, searches the simulation value of the computer system in a systematic way, changes the searched gradient in each iteration calculation until the convergence reaches the waiting time value which accords with the confidence interval. The random gradient search optimization algorithm provided by the invention can find out the approximate optimal solution of the optimization model under the requirements of large solution space, solving time and computer simulation sampling cost.
It should be understood that parts of the specification not set forth in detail are well within the prior art.
It should be understood that the above description of the preferred embodiments is given for clarity and not for any purpose of limitation, and that various changes, substitutions and alterations can be made herein without departing from the spirit and scope of the invention as defined by the appended claims.

Claims (1)

1. A random gradient search optimization method for a security inspection system with a risk screening mechanism is characterized by comprising the following steps:
step 1: constructing a risk checking system database;
step 2: constructing a customs clearance security inspection random restriction type mixed integer type optimization model;
the customs clearance security inspection random restriction type mixed integer type optimization model is as follows:
maximizing the objective function: d1·R1(τ)+d2·R2(τ)
Satisfies the constraint formula 1: w (τ, s)1,s2)≤ε
The restriction formula 2:
Figure FDA0003594660160000011
restriction formula 3: tau is more than 0 and less than 1
Restricted formula 4: s1,s2∈positive integer
Wherein τ is a risk classification threshold, s, at which clearance passengers are assigned to different security check capability checkpoints1Number of security personnel on duty for conventional security check station, s2The number of security check personnel on duty of the strict security check station; r1(τ) is the security level of the customs passenger at the conventional security checkpoint when the classification threshold is τ; r2(τ) is the level of security of the customs passenger at the strict security checkpoint when the classification threshold is τ; d1Accuracy of the security equipment of a conventional security station, d2Accuracy of security equipment of a security station is critical; w () is expected waiting time of the customs passenger in the security inspection system, and epsilon is a required value of efficiency of the expected waiting time of the customs passenger; p (τ) is the distribution probability that a clearance passenger is assigned to a strict security check station when the classification threshold is τ; beta is a1Depreciation and depreciation cost beta for security inspection equipment of each year conventional security inspection station2Is strict every yearDepreciating and depreciating the cost of security inspection equipment of the security inspection station; c. C1The duty cost of the security personnel in the conventional security inspection station, c2The duty cost of security personnel in a strict security station; b is the total budget amount for building a security check station and hiring a security check worker; the positive integer is a positive integer;
and step 3: solving the optimal risk threshold value tau of the customs clearance security inspection random restricted mixed integer type optimization model by utilizing a random gradient search optimization algorithm*And number of security personnel
Figure FDA0003594660160000012
The specific implementation of the step 3 comprises the following substeps:
step 3.1: let K be the number of iterations, K the maximum number of iterations,
Figure FDA0003594660160000013
for the upper bound of the risk classification gate threshold,τthe lower bound of the threshold of the risk classification gate, M is the number of repetitions of the simulation test, θ1And theta2Improving the gradient parameters of the solution for each iteration;
step 3.2: according to the total budget limit of customs
Figure FDA0003594660160000014
Selecting s1And s2Set of all possible combinations of
Figure FDA0003594660160000015
Step 3.3: setting the iteration number n of the optimization solution to be 0, and starting to execute the following iteration;
step 3.3.1: from the collection
Figure FDA0003594660160000021
Get a set of solutions
Figure FDA0003594660160000022
Setting the iteration number k of the algorithm to 0And set τ1＝τ0；
Step 3.3.2: set k ← k +1 and
Figure FDA0003594660160000023
according to τkPerforming M repeated computer simulation tests, and recording the obtained clearance time
Figure FDA0003594660160000024
And calculating the average value thereof
Figure FDA0003594660160000025
And
Figure FDA0003594660160000026
95% confidence interval of
Figure FDA0003594660160000027
Wherein γ is the half-length of the confidence interval;
step 3.3.3:
if ε satisfies the inequality
Figure FDA0003594660160000028
Or when the number of algorithm iterations K equals K, then set
Figure FDA0003594660160000029
And update the set
Figure FDA00035946601600000210
Otherwise, it is set
Figure FDA00035946601600000211
Wherein theta is1When the gate valve value epsilon of the expected closing time limiting type is larger than the upper limit value of the confidence interval, the gradient parameter is used; and theta2When the threshold value epsilon is smaller than the upper limit value of the confidence interval, the gradient parameter is used; and rotating to execute the step 3.3.2;
step 3.3.4: when the collection
Figure FDA00035946601600000212
When the algorithm is started, the loop of the main program of the random gradient search optimization algorithm is ended and the step 4 is entered; otherwise, setting an iteration number n ← n +1 of the optimized solution, and performing step 3.3.1 in a rotating manner;
step 3.4: optimizing the algorithm's resulting solution for random gradient search
Figure FDA00035946601600000213
Verifying the feasibility of the program judgment solution by using the feasibility in the sequencing and selecting program, and deleting the infeasible solution under the statistical guarantee; order to
Figure FDA00035946601600000214
All selected by the feasibility verification program are shown
Figure FDA00035946601600000215
Forming a solution set;
step 3.5: obtaining an optimal risk gate valve value
Figure FDA00035946601600000216
And the optimal safety level of the customs clearance security inspection system is SL (tau)*)；
And 4, step 4: carrying out risk judgment on the passenger to be checked and the carry-on luggage goods by utilizing a risk checking system database;
if the risk value of the passenger to be checked and the carry-on luggage goods is less than the risk threshold value tau*Making the mobile phone go to a 'conventional inspection station' for safety inspection; otherwise, if the risk value of the passenger to be checked and the carry-on luggage cargo is more than or equal to the risk threshold tau*And then go to a strict security check station for security check.
CN201910486476.3A 2019-06-05 2019-06-05 Random gradient search optimization method for security inspection system with risk screening mechanism Expired - Fee Related CN110276112B (en)

Priority Applications (1)

Application Number Priority Date Filing Date Title
CN201910486476.3A CN110276112B (en) 2019-06-05 2019-06-05 Random gradient search optimization method for security inspection system with risk screening mechanism

Applications Claiming Priority (1)

Application Number Priority Date Filing Date Title
CN201910486476.3A CN110276112B (en) 2019-06-05 2019-06-05 Random gradient search optimization method for security inspection system with risk screening mechanism

Publications (2)

Publication Number Publication Date
CN110276112A CN110276112A (en) 2019-09-24
CN110276112B true CN110276112B (en) 2022-06-07

Family

ID=67962031

Family Applications (1)

Application Number Title Priority Date Filing Date
CN201910486476.3A Expired - Fee Related CN110276112B (en) 2019-06-05 2019-06-05 Random gradient search optimization method for security inspection system with risk screening mechanism

Country Status (1)

Country Link
CN (1) CN110276112B (en)

Families Citing this family (2)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
CN111352171B (en) * 2020-03-30 2023-01-24 重庆特斯联智慧科技股份有限公司 Method and system for realizing artificial intelligence regional shielding security inspection
CN111784131A (en) * 2020-06-19 2020-10-16 江苏金匮通供应链管理有限公司 Customs processing method and system based on customs data analysis

Citations (4)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
JP2006039987A (en) * 2004-07-28 2006-02-09 Mitsubishi Heavy Ind Ltd Check system calculating device and check system calculation method
CN104063816A (en) * 2014-06-30 2014-09-24 中国民航机场建设集团公司 Method for security check of airport passenger checked baggage
CN108805344A (en) * 2018-05-29 2018-11-13 五邑大学 A kind of high-speed railway network train running scheme optimization method considering time-dependent demand
CN109784819A (en) * 2019-03-19 2019-05-21 东部机场集团有限公司 Shipping safety check classification hierarchy system and its stage division

Patent Citations (4)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
JP2006039987A (en) * 2004-07-28 2006-02-09 Mitsubishi Heavy Ind Ltd Check system calculating device and check system calculation method
CN104063816A (en) * 2014-06-30 2014-09-24 中国民航机场建设集团公司 Method for security check of airport passenger checked baggage
CN108805344A (en) * 2018-05-29 2018-11-13 五邑大学 A kind of high-speed railway network train running scheme optimization method considering time-dependent demand
CN109784819A (en) * 2019-03-19 2019-05-21 东部机场集团有限公司 Shipping safety check classification hierarchy system and its stage division

Non-Patent Citations (1)

* Cited by examiner, † Cited by third party
Title
基于旅客分类的机场安检系统研究;赵振武等;《综合运输》;20170315(第03期);全文 *

Also Published As

Publication number Publication date
CN110276112A (en) 2019-09-24

Similar Documents

Publication Publication Date Title
Benabdellah et al. A survey of clustering algorithms for an industrial context
CN110503245B (en) Prediction method for large-area delay risk of airport flight
CN102637143B (en) Software defect priority prediction method based on improved support vector machine
CN110276112B (en) Random gradient search optimization method for security inspection system with risk screening mechanism
US20220012512A1 (en) Intelligent gallery management for biometrics
CN107786369A (en) Based on the perception of IRT step analyses and LSTM powerline network security postures and Forecasting Methodology
CN108304974A (en) A kind of civil aviaton NOSHOW predictions based on optimization C5.0 and Apriori and strong factor-analysis approach
Liu et al. A novel CBR system for numeric prediction
CN111985685A (en) Flight plan optimization method and device based on big data analysis and electronic equipment
Gorodnichy et al. Automated border control: Problem formalization
CN111950910A (en) Airport guarantee vehicle task scheduling method based on DBSCAN-GA
CN111079827B (en) Railway data state evaluation method and system
Pugnana et al. AUC-based selective classification
Papagiannopoulos et al. Understanding and predicting passenger behaviours through data analytics
CN112232652A (en) Passenger risk level classification method and device, electronic equipment and storage medium
Li et al. Wind pressure coefficients zoning method based on an unsupervised learning algorithm
Tsai et al. Simulation optimization in security screening systems subject to budget and waiting time constraints
CN111209955A (en) Airplane power supply system fault identification method based on deep neural network and random forest
CN114882974B (en) Psychological diagnosis database access artificial intelligence verification system and method
CN110555604A (en) Laboratory comprehensive evaluation system
Lasrado et al. Whose Maturity is it Anyway?: The Influence of Different Quantitative Methods on the Design and Assessment of Maturity Models
Lee et al. Protecting air transportation: a survey of operations research applications to aviation security
CN115345408A (en) Terminal area approach efficiency assessment method based on entropy weight method
Has et al. Modelling energy efficiency of public buildings by neural networks and its economic implications
Xu et al. Short-term Passenger Flow Forecasting of the Airport Based on Deep Learning Spatial-temporal Network

Legal Events

Date Code Title Description
PB01 Publication
PB01 Publication
SE01 Entry into force of request for substantive examination
SE01 Entry into force of request for substantive examination
GR01 Patent grant
GR01 Patent grant
CF01 Termination of patent right due to non-payment of annual fee

Granted publication date: 20220607