JP4959599B2 - Method and apparatus for dynamic navigation of operator operation - Google Patents

Description

  The present invention relates to an execution order of operator operations to be input / changed in a predetermined application program (business application) in which an operator inputs / changes values of a plurality of types of data items by interactive processing via a computer terminal. The present invention relates to a technology that appropriately presents and restricts during processing.

  Companies are trying to increase business efficiency through IT as business operations become more complex. Business can be defined in terms of “who performs what operation in what order”. In general, these rules were established as a business manual, and the operator proceeded with the work in accordance with this rule. However, the complexity of this code, such as the expansion of business, compells operators to become familiar with the business manual, and the employment situation (difficulty in securing high-quality human resources, shortening of training and employment periods, etc.) Considering this, it becomes a major obstacle to corporate activities.

  In order to solve this, it is conceivable that the business application is responsible for “management of execution authority for operations”, “(semi) automation of operations” and “control of the order of operations”. When an operator proceeds with a business using such a business application, the business application sequentially presents executable operations, and further automatically processes each operation as much as possible. By constructing an interactive business application in this way, an operator with less business knowledge can efficiently proceed with business according to the presentation of the business application.

<Conventional technology>
A flexible process control engine has been proposed as one of tools for developing a business application that performs the interactive processing (see Non-Patent Documents 1 and 2). One of the features of the flexible process control engine is that “control of the order of operations” is realized by dynamic navigation. Dynamic navigation refers to presenting an operator operation candidate to be executed next to an operator according to the progress of the current job in order to complete the job.

  The flexible process control engine holds a problem solver called a planner to realize this dynamic navigation. The planner is a solution search computer program for a search problem called a planning problem in the field of artificial intelligence (see Non-Patent Document 3).

  When using the planner, the preconditions for executing each operation on the state space of the target area of the business application and the effect on the state after the execution of the operation are defined in advance, and the initial target area Specify the status (current status of business progress) and the target status (business completion status). A set of this set of operations (preconditions / effects), initial state, and target state is called a planning problem. When these are given to the planner, the operation order (plan) from the initial state to the target state can be obtained.

  The flexible process control engine has a function of using a planner to obtain all operations that can reach the target state from among the executable operations whose preconditions match the initial state. The operation obtained by this function is called an operation candidate.

  If a business application is developed using the function for obtaining the operation candidate, the dynamic navigation for the operator can be easily realized. Specifically, dynamic navigation can be realized by the following processing method.

  (1) Notify the flexible process control engine of the status of the business application changed by the actual operation performed by the operator.

  (2) Let the flexible process control engine obtain a set of operation candidates in the notified state.

  (3) Present the obtained set of operation candidates to the operator, and select and execute an operation to be actually performed from the set.

  The processes from (1) to (3) are repeated until the end of business.

At this time, the function (2) for obtaining a set of operation candidates is realized by solving the planning problem a plurality of times (see FIG. 2 of Non-Patent Document 2). In order to make a decision, it is necessary to check that the plan exists in the state after the operation is executed, so the set of operation candidates is planned for the number of operations that can be executed from the state notified in (1). It can be obtained by solving the problem.
Nosue, Nakajima, Oishi, Masuda, Yamamura, “Control Technology for Planning Diversity Using Constraints and Satisfaction”, IEICE 43rd Telecommunication Management Study Group, IEICE Technical Report, Vol. 106, no. 600, 2006, pp. 71-76, H. Nozue, H. Nakajima, H. Oishi, T. Masuda, and T. Yamamura, "Process Control Technique Using Planning and Constraint Satisfaction", In Proc. Of the 10th Asia-Pacific Network Operations and Management Symposium (APNOMS 2007) , LNCS, Vol.4773, 2007, pp. 62-71. Internet <URL: http://www.cs.rochester.edu/u/kautz/satplanfblackbox/index.html>

  However, the above processing method has a problem regarding the execution speed of the business application.

  As described above, a set of operation candidates required by the current flexible process control engine is realized by obtaining a plan by a planner. It is known that the amount of calculation for obtaining this plan is proportional to the number of states in the target area. In the above processing method, when the operator uses a business application, the process (2) is executed for each operation. For this reason, in a business application that targets a large-scale business, it takes time to respond until an operation candidate is presented to the operator after the operator performs an operation, and the responsiveness deteriorates.

  Therefore, in the present invention, by obtaining a set of reachable states to the target state as pre-processing, it is possible to obtain operation candidates with a logarithmic order calculation amount with respect to the number of business application states during execution of the business application. A dynamic navigation processing method is realized.

  The “reachable state” is a state where the target state can be reached. Therefore, if a certain state is a reachable state, there is always a plan for reaching the target state from that state. As shown in FIG. 1, the planning problem handled by the flexible process control engine can be considered as a problem of obtaining a transition sequence from an initial state to a target state in a finite state transition system. The reachable state in FIG. 1 is shown in FIG.

  The fact that a certain state is a reachable state is synonymous with the existence of a plan from that state to the target state. Therefore, in the process of obtaining a set of operation candidates in the prior art, if the part that solved the planning problem is to obtain a set of reachable states in advance, whether or not the state is included in the set of reachable states. It can be replaced with a process for determining.

  A method for obtaining and presenting the execution order of operator operations using a computer using this method will be described with reference to FIGS.

  First, before executing the business application, the reachable state derivation step for obtaining a set of reachable states to the target state as preprocessing is executed as shown in FIG.

  When a set of planning problem operations (preconditions / effects) and a target state defined in advance for a business application are input from the input device to the computer (FIG. 3; s1), a set R of reachable states is obtained from these. Is obtained (FIG. 3; s2), and is output and stored in the storage device (FIG. 3; s3).

  Fig. 3; The technology for obtaining the set R of reachable states of s2 includes JR Burch, EM Clarke, KL McMillan, DL Dill, and J. Hwang, "Symbolic model checking: 10E20 states and beyond", In Proc. of the International Conference on Logic in Computer. Science, 1990, etc. are known and can be used. The processing in FIG. 3 may be performed only once unless the business application is changed, that is, the set of operations (preconditions / effects) and the target state are not changed.

  Next, an operation candidate presentation step for obtaining a set of operation candidates to be presented to the operator during execution of the business application is executed as shown in FIG.

  First, a (current) state s of a business application and a set of planning problem operations O defined in advance for the business application are input from the input device to the computer (FIG. 4; s11). The operation candidate set C to be executed is set to an empty set (FIG. 4; s12).

  Next, one operation is taken out from the operation set O (FIG. 4; s13), and it is determined whether or not the operation o can be executed in the state s (FIG. 4; s14). If it cannot be executed, the process returns to the processing of s13. If it is executable, the state s ′ after the operation o is obtained from the state s (FIG. 4; s15), and the state s ′ is stored in the storage device. It is checked whether it is included in the set R of reachable states (FIG. 4; s16). If it is not included, the process returns to the processing of s13, and if it is included, the operation o is added to the operation candidate set C (FIG. 4; s17).

  The processing from FIG. 4; s13 to FIG. 4; s17 is performed for all the operations, that is, until the set obtained by removing the operation o from the set of operations O becomes an empty set (FIG. 4; s18). The candidate set C is output and displayed on the display device (FIG. 4; s19).

  In the prior art, the process of FIG. 4; s16 is a process for obtaining a plan, and this process requires a calculation amount proportional to the number of states. However, the determination that the state s ′ is included in the set R of reachable states of the present invention is performed by using the binary decision tree or the Ordered Binary Decision Diagram (RE Bryant, “Graph-based algorithms for” Boolean functions manipulation ", IEEE Transactions on Computers, C-35 (8), pp.677-691. Possible approaches are known. As a result, application responsiveness can be improved and dynamic navigation can be performed as compared with the prior art.

  According to the present invention, in a business application constructed using a flexible process control technology, the process navigation being executed can be performed at a higher speed than the conventional technology, and the responsiveness of the business application can be improved. .

  FIG. 5 shows an example of an apparatus for carrying out the dynamic navigation method of the operator operation according to the present invention, here an example realized on a known computer. In the figure, 1 is an input device, 2 is a display device, 3 is It is a computer.

  The input device 1 includes a keyboard, a mouse, and the like, and is used for inputting predetermined information, in this case, a planning problem and a state of a business application, to the computer 3. The display device 2 includes an image display device such as a liquid crystal display and a CRT, and is for presenting a predetermined processing result of the computer 3, here, a set of operation candidates.

  The computer 3 controls each device according to a predetermined program corresponding to the processing flow shown in FIG. 3 to FIG. 4, and configures the storage device 4, the preprocessing device 5, and the operation candidate presentation device 6 at this time.

  The storage device 4 stores a set of reachable states output from the preprocessing device 5 and presents the information in response to a request from the operation candidate presentation device 6.

  In accordance with the reachability state derivation step shown in FIG. 3, the preprocessing device 5 includes a set of planning problem operations (preconditions / effects) defined in advance for the business application input from the input device 1 and the target state. A set of reachable states is obtained from the data, and is output to the storage device 4 for storage.

  The operation candidate presentation device 6 refers to the current state of the business application input from the input device 1 and the set of reachable states stored in the storage device 4 according to the operation candidate presentation step shown in FIG. Then, a set of operation candidates to be executed next is obtained from a set of operations (preconditions / effects) of planning problems defined in advance for the business application input from the input device 1 and output to the display device 2 To display.

  With this apparatus, in a business application constructed by flexible process control technology, the process navigation being executed can be performed at a higher speed than the conventional technology, and the responsiveness of the business application can be improved.

  In the above-described embodiment, an example in which the storage device 4, the preprocessing device 5, and the operation candidate presentation device 6 are realized on a computer has been shown, but may be realized by hardware. Each of these devices exists in a distributed manner on a plurality of pieces of hardware connected to each other via an appropriate communication path, and can be executed while communicating with each other.

  An example of dynamic navigation according to the present invention will be described.

  First, assume that the following planning problem is defined for a business application. However, here, it is assumed that the state of the business application is expressed by a set of In-1, In-2, and In-3 terms.

initial state {}
Target state {In-1, In-2, In-3}
Operation act1 (no precondition, effect In-1)
act2 (precondition ¬In-3, effect In-2)
act3 (Precondition In-1, Effect In-3)
act4 (Preconditions In-2, In-3, Effect In-1)
“¬” represents “Negation”.

An operation o can be performed on the (current) state s.
● All terms appearing in the preconditions for operation o are elements of the set of terms in state s.
And,
All negative terms appearing in the preconditions of operation o are not elements of the set of terms in state s,
If the condition is met.

  The former condition means that, for example, the operation act3 can be executed in the state {In-1}, but the operation act4 cannot be executed. The latter condition means that the operation act2 can be executed in the state {In-1}, and the operation act2 cannot be executed in the state {In-3}.

  When the operation o can be executed from a certain state s, the state s can be updated based on the effect of the operation o. Here, the state is updated as follows.

Add all terms that appear in the effect of operation o to the set of terms in state s.
Remove all negative terms that appear in the effect of operation o from the set of terms in state s.

  In the former update, for example, the state after executing the operation act2 from the state {In-1} means that the effect In-2 is added and becomes {In-1, In-2}. On the other hand, the latter means that when an operation having an effect ¬In-1 is executed from the state {In-1}, the state becomes {}.

  Under this definition, the state transition system of the planning problem is as shown in FIG.

  In the present invention, before execution of a business application, for example, when a business application is created, a set of operations of the planning problem and a target state are input from the input device 1, and the set of reachable states shown in FIG. It is obtained and stored in the storage device 4.

  Next, an example of dynamic navigation when an operator actually operates a business application, that is, an operation candidate set is presented.

  For example, it is assumed that the current state of the business application is the initial state {} in FIG. 1 and the operator has performed an operation corresponding to act1 on the business application from that state. Then, the state of the business application transitions to {In-1}. At this time, the business application inputs the state {In-1} after the transition and a set of planning problem operations from the input device 1 of the present invention in response to the state transition. Thereby, the operation candidate presentation device 6 of the present invention sets an empty set as an initial value of a set of operation candidates to be obtained in accordance with the operation candidate presentation step of FIG. 4 (FIG. 4; s12).

  Next, one of the input operations act1, act2, act3, and act4, for example, the operation act2 is taken out, and it is determined whether the operation act2 can be executed (FIG. 4; s13, s14).

  Since the precondition of act2 can be executed at ¬In-3, the state {In-1, In-2} after execution of the operation act2 is obtained (FIG. 4; s15). Then, with reference to the set of reachable states stored in the storage device 4 (FIG. 2), it is determined whether or not the state {In-1, In-2} is included in the set (FIG. 4; s16). .

  Since this state is now included in the set of reachable states, operation act2 is added to the set of operation candidates (FIG. 4; s17). This is executed for the remaining operations act1, act3, and act4. In this case, there is no addition to the operation candidate set other than the operation act2, and as a result, the operation candidate set becomes {act2}, which is output to the display device 1 and displayed.

  In this way, the operator can select an operation from the displayed set of operation candidates and execute the business application.

Explanatory diagram showing an example of a state transition system for planning problems Explanatory drawing which shows the reachable state in the state transition system of FIG. The figure which shows the processing flow of the reachable state derivation | leading-out step of the dynamic navigation method of operator operation of this invention The figure which shows the processing flow of the operation candidate presentation step of the dynamic navigation method of operator operation of this invention The block diagram which shows an example of embodiment of the dynamic navigation apparatus of the operator operation of this invention

Explanation of symbols

  1: input device, 2: display device, 3: computer, 4: storage device, 5: preprocessing device, 6: operation candidate presentation device.

Claims (4)

Display of execution order of operator operations on a predetermined application program operating on an arbitrary computer on the arbitrary computer or another computer connected to the arbitrary computer via a network during the execution of the predetermined application program A method of presenting to a device,
The arbitrary computer or another computer connected to the arbitrary computer via a network,
A reachable state derivation step of obtaining a set of reachable states from a set of operations of the predetermined application program input from an input device and a target state in the predetermined application program , and outputting and storing the set in a storage device;
With reference to the current state of the predetermined application program input from the input device and the set of reachable states stored in the storage device, from the set of operations of the predetermined application program input from the input device An operation candidate presenting step for obtaining a set of operation candidates to be executed next having only an operation that can reach the target state in the current state as an element, and outputting and displaying the operation candidate on a display device. A dynamic navigation method of operator operation. The operation candidate presentation step according to claim 1 comprises:
A first step in which a current state of a predetermined application program and a set of operations are input from an input device;
A second step of setting a set of operation candidates to be executed next to an empty set;
A third step of taking one operation from the set of operations and determining whether it can be performed from the current state;
A fourth step for obtaining a post-execution state for an executable operation based on the effect of the operation;
A fifth step of determining whether the state after execution is included in a set of reachable states stored in a storage device;
A dynamic navigation method for operator operations, comprising at least a sixth step of adding the extracted operation to a set of operation candidates when the state after execution is included in a set of reachable states.   The program for making a computer perform each step of Claim 1 or 2. Display of execution order of operator operations on a predetermined application program operating on an arbitrary computer on the arbitrary computer or another computer connected to the arbitrary computer via a network during the execution of the predetermined application program A device presenting to the device,
A next device provided on the arbitrary computer or another computer connected to the arbitrary computer via a network, that is, a storage device for storing a set of reachable states;
A preprocessing device that obtains a set of reachable states from a set of operations of the predetermined application program input from an input device and a target state in the predetermined application program , and outputs and stores the set in a storage device;
With reference to the current state of the predetermined application program input from the input device and the set of reachable states stored in the storage device, from the set of operations of the predetermined application program input from the input device An operation candidate presenting device that obtains a set of operation candidates to be executed next, using only operations that can reach the target state in the current state as elements, and outputs and displays them on a display device. A dynamic navigation device for operator operation.

Images