JP5664494B2 - Planning support apparatus, planning support method, and computer program - Google Patents

Description

  The present invention relates to a planning support apparatus, a planning support method, and a computer program that support planning of arrangement of an object in a predetermined arrangement area.

  Arrangement plan for arranging objects in a predetermined arrangement area, such as determining the arrangement layout of exhibits in museums, planning the order of production of products in the manufacturing industry, organizing lesson schedules in schools, etc. Planning is done in various situations.

  For example, in a museum, planned exhibitions are regularly held, and the person in charge determines the arrangement of exhibits to be exhibited in the exhibition room in consideration of the legibility of the viewer, the intention of the planner, and the like. For example, as shown in FIG. 17, first, an arrangement area 12 in which an exhibit can be arranged in the exhibition room 10 is determined based on the size of a desk on which the exhibit is placed, the size of an explanation board describing the contents of the exhibit, and the like. Is set. Thereafter, the exhibits 20 (A to R) are arranged in the set arrangement region 12. In the exhibition room 10 shown in FIG. 17, three arrangement regions 12 are arranged adjacent to each other so that the exhibits 20 are displayed in groups of three. FIG. 18 shows an arrangement example in which the exhibits 20 are arranged in the arrangement region 12 of the exhibition room 10 shown in FIG.

  Similarly, as an example of planning the product manufacturing order in the manufacturing industry, the business of making a casting plan in the steel industry can be considered in the same manner as the determination of the layout of the exhibits in the museum. For example, as shown in FIG. 19, a charge 40 that needs to be produced from the number of charges per day that can be cast by a continuous caster and corresponding to the output steel frame 32 is cast. (Corresponding to the object) is arranged. The symbols “A” to “P” attached to each charge 40 in FIG. 19 are symbols representing the steel type of the charge. In the example shown in FIG. 19, there are 6 sets of steel-out frames, but the close charge means that continuous casting is performed, and continuous casting is terminated at a portion where there is an interval, and the setup change is performed. Means to do. The casting lot 30 from the start to the end of continuous casting is called casting. In FIG. 19, casting of 2 casts per day (6 × 2 = 12 charges) is performed.

  The work of organizing the lesson schedule can be basically considered in the same manner as the work of determining the layout of the exhibits in the museum and making a casting plan for the steel industry. Class schedules are organized by arranging class content (corresponding to objects) such as national language and arithmetic, etc., for class frames (corresponding to the arrangement area) for each day of the week.

  As described above, in many organizations, an operation of arranging an object in a predetermined arrangement area is performed. At this time, the number of arrangement areas and the number of objects do not need to be the same, and the number of arrangement areas and the number of objects are the same, or the number of objects is larger than the number of arrangement areas. Also, as shown in FIGS. 17 and 19, there are many cases where a plurality of arrangement areas are in one set, and here, a group consisting of a plurality of arrangement areas is described as a lot.

  When placing an object in the placement area, it is necessary to consider a placement that is as good as possible while considering the placement constraints of the object. Here, the placement constraint is a requirement that must be satisfied when placing an object in the placement area. Broadly speaking, there are a restriction on the placement location of the object and a restriction on the context with other objects. .

  For example, restrictions on the location of the target object include the requirement that “a target object must not be (or must be) placed at the beginning (or end) of a lot” or “ There is a requirement that “the object is arranged so that the set fitness is as high as possible”. In addition, the fitness set for the object can be determined according to the route in the determination of the layout of the exhibits in the museum, for example, and in the work of planning the steel industry casting plan, the order of casting based on the delivery date Can be determined accordingly.

  On the other hand, as a restriction on the context with other objects, for example, a requirement that “the object A and the object B should not be placed close to each other” or “the object C and the object D are There is a requirement that “place as close as possible”. In addition, there is a requirement that “a plurality of objects of the same type do not exist in the same lot”, a requirement that “objects of the same type are arranged close to each other”, and the like.

  The person in charge who performs the arrangement work of the object arranges the object in the arrangement area in consideration of the above arrangement restrictions. At this time, if the number of arrangement areas and objects is small, the burden on the person in charge is small even if all work is done manually, but if the number of arrangement areas and objects is large and there are many arrangement restrictions, the burden on the person in charge also increases. In view of this, a method for reducing the burden on the person in charge of the object placement work has been proposed.

  For example, a method using an expert system has been proposed. An expert system is a software program having the ability to give a computer expert knowledge about a specific area and solve a problem based on the expert knowledge (Non-Patent Document 1). Expert systems have been introduced for various problems ranging from mechanical system fault diagnosis and computer system design to large-scale complex work schedules. However, when using an expert system, even if there is no clear guideline, the expert determines the placement of the object on a trial and error basis, or the difference between the skilled and non-skilled personnel is clearly different. In many cases, it is difficult to realize an expert system.

  As an approach different from the expert system, the placement constraints are expressed as mathematical formulas and evaluation functions, and using the methods such as mathematical programming and genetic algorithm, the placement of the target with the best evaluation value as the output value of the evaluation function A method for automatically calculating a plan has also been proposed (Non-Patent Documents 2 and 3). However, in this case, it is difficult to determine an evaluation function representing the quality of the object placement plan for the placement region, and it is difficult to propose a placement plan that can be satisfied even by a skilled person.

  For example, in determining the museum's exhibition layout, there are evaluation items based on sensibilities that cannot be expressed in words, such as a sense of unity, as well as constraints such as the placement location of exhibits, etc. There are many. Therefore, when deciding the actual display layout, the evaluation items are programmed even if the ideal layout does not match among the staff in charge, the display manager, the designer, etc. It is difficult to present the object arrangement plan reflecting the preference of the person in charge by the system.

  Therefore, a technique has been proposed in which the preferences of the user who is a user of the system are modeled and information is presented to the user based on the model (Non-Patent Documents 4 and 5). Non-Patent Document 4 constructs a user preference model using a Bayesian network for content recommendation in car navigation systems. In this system, content, such as restaurants and music, is calculated by probabilistic reasoning in consideration of user data and situation data, and is proposed to the user in descending order of content that matches the user's preference. . Further, Non-Patent Document 5 implements and evaluates a movie recommendation system for mobile terminals that considers both individuality and situation. In this research, we have developed a system for recommending movies that are conscious of their use on mobile terminals, taking into account individual differences in user preferences and situational dependency at the same time. It is suggested that the user's preference for movies can be predicted with higher accuracy by taking the user's situation into consideration, and the satisfaction of the recommendation list can be improved.

  However, in the techniques in Non-Patent Documents 4 and 5, since the user situation is used to recommend content suitable for the user, it is necessary for the user to explicitly input the user situation every time the system is used.

  On the other hand, an apparatus that automatically determines a user situation from an operation history and proposes an alternative has also been proposed (Patent Documents 1 and 2). In Patent Document 1, when a user inputs a destination, the user's ambiguity is estimated from the time interval required for inputting each character, and other destinations in the same genre as the destination are determined based on the ambiguity. A proposed navigation device is disclosed. Further, Patent Document 2 proposes a recommended destination candidate near the current location of the vehicle, then estimates whether or not the driver intends to go to the proposed destination candidate from the driving behavior information, and depending on the presence or absence of the road An information presentation system for changing the content of guidance is disclosed.

JP 2010-127676 A JP 2011-99747 A

Mitsuaki Higuchi, Morio Nagata, Application of Genetic Algorithm to Flight Scheduling Problem, IEICE Technical Report, AI, Artificial Intelligence and Knowledge Processing, Vol. 95, No. 572, pp.9-14, 1996-03 -13. Tezuka Univ., Masahiro Kaji, Masaharu Munasa, Kiyoshi Akama, Risk Visualization and Decision Support for Supply Planning under Uncertain Demand (Social / Human Information Systems, Special Feature: Information Systems that Open up New Application Areas) ), IPSJ Journal, Vol. 47, No. 3, pp. 701-710, 2006-03-15. Ikuo Hatano, Junichi Nishiyama, Motohide Umano, Tanyuki Tamura, Development of distributed simulation system for production system to support real-time scheduling system construction (simulation), Symposium, No. 32, pp. 96-99, 1994-10-11. Hirotoshi Iwasaki, Nobuhiro Mizuno, Kosuke Hara, Yoichi Motomura, Application of Bayesian network to car navigation system recommending contents according to user's preference, IEICE Technical Report, NC2004-55, pp. 25-30, 2004. Tomohiro Ono, Shigeaki Kurokawa, Yoichi Motomura, Hideki Aso, Realization and Evaluation of Movie Recommendation System Considering Individual and User Differences in User Preferences, IPSJ Transactions, Vol. 49, No. 1, pp. 130- 140, 2008-01-15. Kazuo Shigeru, Masaomi Ueno, Yoichi Motomura, Bayesian network overview, Baifukan

  However, in order to apply the techniques described in Patent Documents 1 and 2 to an arrangement task in which it is difficult to formalize evaluation items, a clear planning guideline such as a destination cannot be obtained from the user. Even if the degree of hesitation can be estimated from this, an effective alternative cannot be proposed. Moreover, even if an alternative is proposed to the dark clouds as in Patent Document 2, in many cases, it is unnecessary information for the user and is annoying. Furthermore, the alternative expected by the user differs between when the object is hardly arranged in the arrangement area and when the object is arranged in all the arrangement areas. For example, when almost no object is placed in the placement area, a placement plan that is completely different from the user's placement plan is helpful for the user. On the other hand, after the objects have been placed in all the placement areas, even if a placement plan that is completely different from the user's placement plan is proposed, it is not very helpful for the user, but rather a fine correction plan that is close to the user's placement plan is presented. This is useful information for the user.

  Thus, with the conventional method, an effective alternative according to the user's internal situation cannot be presented from the system, and the burden on the user cannot be reduced sufficiently. There are various burdens on the user. For example, the user may be at a loss in selecting the next object to be placed from among unplaced objects, or may be at a loss in the placement destination of the selected object. In addition, even after all objects have been placed in the placement area, they are unable to have the placement themselves and are worried that there is a better placement, and often waste time by repeating trial and error. .

  Therefore, the present invention has been made in view of the above problems, and an object of the present invention is to provide a novel and effective alternative that can propose an effective alternative according to the situation of the user's arrangement work. An object of the present invention is to provide an improved planning support apparatus, a planning support method, and a computer program.

In order to solve the above problems, according to an aspect of the present invention, a planning support apparatus that supports planning of the arrangement of a plurality of objects in a plurality of arrangement areas is provided. The planning support apparatus includes a progress degree estimation unit that estimates a progress degree that indicates the progress of the placement work based on a planned filling rate that represents a ratio of the target object that is placed with respect to the placement area, and a target object with respect to the placement area. Based on the operation history of the user who arranges the placement area based on the degree of progress and the degree of doubt And a proposal mode determining unit that determines a proposal mode that represents the feature of the arrangement plan for the object, and an alternative creation unit that creates an alternative to the arrangement plan based on the proposal mode.

  The planning support apparatus of the present invention estimates the progress of the placement work and the user's lost state in the placement work from the situation of the user's placement work, and creates a placement alternative in a proposal mode according to the user's situation. . Thereby, it is possible to propose an effective alternative according to the situation of the user's arrangement work.

  The proposal mode determination unit is arranged based on the alternative adjustment information that represents the selectivity, which is a ratio of the proposal mode for the planning situation calculated from the progress and the degree of ambiguity, which is generated in advance based on the operation history. The degree of selectivity of the proposal mode corresponding to the progress level and the degree of ambiguity during work may be calculated, and one proposal mode may be determined from a plurality of proposal modes according to the degree of selectivity.

  Here, the alternative adjustment information is information that expresses the selectivity of the proposal mode with respect to the user's planning status calculated from the degree of progress and the degree of ambiguity as a conditional probability, and can be acquired using Bayesian estimation.

  In addition, the progress degree estimation unit is based on the planned filling rate, until the arrangement object is in the arrangement object browsing state in which the object is not arranged in the arrangement area, until all the objects are arranged once in the arrangement area. The frequency of each state of the first draft creation state, which is a state, and the pushing state, which is a state after all the objects have been once disposed with respect to the placement region, may be represented as the degree of progress.

  At this time, the progress degree estimation unit again increases the planned filling rate from the thrust divergence state and the thrust divergence state in which the planned filling rate decreases after all objects are once arranged in the arrangement region as the frequency of the pushing state. The frequency of each state of the thrust convergence state and the thrust determination state in which all objects are arranged may be represented as the progress.

  Further, the wonder level estimation unit may estimate the frequency of the state with a suspicion and the frequency of the state without a suspicion representing the stagnation state of the user to obtain the level of ambiguity.

  Here, the ambiguity estimation unit may estimate the ambiguity based on a plan creation stagnation time that is an elapsed time from the last event for the object in the placement work.

  Alternatively, the degree of ambiguity estimation unit may estimate the degree of ambiguity based on the operation speed for operating the object in the placement work.

  The ambiguity estimation unit may estimate an object ambiguity representing the ambiguity of the object to be arranged with respect to the arrangement area and an installation place ambiguity representing the ambiguity of the arrangement area where the object is arranged.

In order to solve the above problems, according to another aspect of the present invention, a planning support method by a planning support apparatus that supports planning of the arrangement of a plurality of objects in a plurality of arrangement areas is proposed. Planning support method, the progress estimation unit, based on the planned filling ratio that represents the percentage which the object is located with respect to the arrangement region, and the progress estimation step of estimating the progress representing the progress of the placement work, lost Based on the user's operation history of placing the object with respect to the placement area by the degree estimation unit, a stray degree estimation step for estimating a stray degree representing a user's stray state when placing the object in the placement area; A proposal mode determination unit determines a proposal mode that represents a feature of a placement plan of an object with respect to the placement area based on the degree of progress and the degree of hesitation, and an alternative creation unit determines a proposal mode based on the proposal mode. And an alternative creation step of creating an arrangement alternative.

Furthermore, in order to solve the above problem, according to another aspect of the present invention, the computer is caused to function as a planning support apparatus that supports planning of the arrangement of a plurality of objects in a plurality of arrangement areas. A computer program is provided. Such a program includes a progress estimation unit that estimates the progress indicating the progress of the placement work based on the planned filling rate that represents the ratio of the target placed with respect to the placement area, and the target for the placement area. Based on the operation history of the user who places the object, based on the degree of progress and the degree of ambiguity, the degree of ambiguity estimating unit that estimates the degree of ambiguity representing the user's confused state when placing the object in the arrangement area, A computer is caused to function as a planning support device including a proposal mode determination unit that determines a proposal mode that represents the characteristics of a placement plan for an object, and an alternative creation unit that creates an alternative plan for the placement plan based on the proposal mode. It is characterized by.

  As described above, according to the present invention, it is possible to provide a planning support apparatus, a planning support method, and a computer program capable of proposing an effective alternative according to the situation of the user's arrangement work. Thus, it is possible to reduce the user load in the arrangement work of arranging the object in the arrangement area.

It is explanatory drawing which shows the outline | summary of a structure of the planning assistance apparatus which concerns on embodiment of this invention. It is explanatory drawing which shows the outline | summary of the function of the planning assistance apparatus which concerns on the embodiment. It is explanatory drawing which shows an example of the input screen in which a user inputs the arrangement plan of the target object with respect to an arrangement | positioning area | region in a planning assistance apparatus, Comprising: An arrangement target object browsing state is shown. It is explanatory drawing which shows an example of an input screen, Comprising: A 1st proposal creation state is shown. It is explanatory drawing which shows an example of an input screen, Comprising: The state by which the target object was arrange | positioned in all the arrangement | positioning areas is shown. It is explanatory drawing which shows an example of the proposal screen which shows an alternative to a user with a planning assistance apparatus. It is a functional block diagram which shows the function structure of the planning assistance apparatus which concerns on the same embodiment. It is explanatory drawing which shows the relationship between a plan filling rate and the progress of arrangement | positioning work. It is explanatory drawing which shows an example of a user condition pattern. It is explanatory drawing which shows an example of proposal mode. It is explanatory drawing which shows an example of alternative proposal adjustment information. It is a flowchart which shows the alternative preparation process by the planning assistance apparatus which concerns on the embodiment. It is a flowchart which shows the alternative preparation process based on the determined proposal mode. It is a block diagram which shows the hardware structural example of the planning assistance apparatus which concerns on the same embodiment. It is explanatory drawing which shows the data of the display target object used for the effect experiment. It is a graph which shows the experimental result of the effect experiment regarding a user's confidence degree and the contribution degree of a planning assistance apparatus. It is explanatory drawing explaining the arrangement layout of the exhibit in an exhibition room. It is explanatory drawing which shows one example of arrangement | positioning which has arrange | positioned the exhibit in the exhibition room shown in FIG. It is explanatory drawing explaining the operation | work which draws up the casting plan of the steel industry.

  Exemplary embodiments of the present invention will be described below in detail with reference to the accompanying drawings. In addition, in this specification and drawing, about the component which has the substantially same function structure, duplication description is abbreviate | omitted by attaching | subjecting the same code | symbol.

<1. Outline of Planning Support Device>
[1-1. Outline of configuration of planning support device]
First, the outline of the configuration of the planning support apparatus according to the embodiment of the present invention will be described with reference to FIG. FIG. 1 is an explanatory diagram showing an outline of the configuration of the planning support apparatus according to the present embodiment.

  The planning support apparatus 100 according to the present embodiment is an apparatus that presents an effective alternative according to a user's work situation. As shown in FIG. 1, the planning support apparatus 100 receives an operation input by a GUI (Graphical User Interface) from a user through an interface, and presents an alternative to the user based on a user situation estimated based on the input content. .

  The user can use the GUI to input a user placement plan for an object with respect to the placement area, and can also input an evaluation for an alternative presented from the planning support apparatus 100. The planning support apparatus 100 acquires an operation history of a user's operation input including a user arrangement plan via an interface, converts the operation history into a numerical value (feature amount of the operation history) for estimating the user's internal situation, and outputs the operation history (a feature amount of the operation history) (see FIG. Operation history acquisition function). Then, the user situation is estimated based on the feature value of the operation history acquired at a predetermined timing (user modeling function). The timing for estimating the user situation can be, for example, a timing for proposing an alternative set at a preset time interval, or can be a timing when a request is made by the user. The planning support apparatus 100 selects an alternative proposal mode according to a user situation (a proposal mode determination function). The proposal mode is information representing the feature of the arrangement plan of the object with respect to the arrangement area created by the planning support apparatus 100, and the content of the alternative proposal proposed to the user greatly varies depending on the proposal mode.

  Then, the planning support apparatus 100 creates an alternative to be presented to the user based on the selected proposal mode (alternative plan creation function). The created alternative is presented to the user via the interface. Thus, by repeating a series of steps of creating an alternative according to the user situation, it is possible to present an effective alternative for the user when the alternative is presented. The user devises the user's arrangement plan until referring to the presented alternative, and finally decides one arrangement plan. In addition, the planning support apparatus 100 can acquire an evaluation for the alternative created by the planning support apparatus 100 from the user via the interface, and the estimated user situation and the proposal mode when the alternative is created, the alternative Record the rating for. The recorded value is used when determining the value of the adjustment coefficient (alternative plan adjustment information described later) used when selecting the proposal mode from the user situation.

[1-2. Overview of planning support device functions]
FIG. 2 shows an outline of the functions of the planning support apparatus 100 shown in FIG. FIG. 2 is an explanatory diagram showing an overview of functions of the planning support apparatus 100 according to the present embodiment. As shown in FIG. 2, the planning support apparatus 100 according to the present embodiment functions the operation history acquisition function, and the ratio of the target object buried in the arrangement area from the operation history of the operation input acquired through the interface. The change in the operation speed from the planned filling rate, the plan creation stagnation time that is the time elapsed from the start of the final event, and the operation speed that represents the speed of the placement operation when creating the placement plan of the object for the placement area The value to be represented (operation speed change value) is acquired.

Next, the user modeling function of the planning support apparatus 100 estimates the user situation based on the information acquired by the operation history acquisition function. For example, the progress of the placement work is estimated based on the planned filling rate. In the present embodiment, the state representing the progress of the placement work is classified into the following five. Each state can be represented by a frequency, and the frequency of each state becomes a progress indicating the progress of the placement work.
(1) Arrangement object browsing state: A state in which an arrangement object (hereinafter also simply referred to as “object”) is not arranged with respect to the arrangement area. (2) First draft creation state: arrangement with respect to the arrangement area State until all objects are arranged once (3) Drought divergence state (thrust state (divergence)): State where the planned filling rate is lowered after all the arrangement objects are once arranged in the arrangement region (4) The thrust convergence state (thrust state (convergence)): The state where the planned filling rate rises again from the thrust divergence state (5) The thrust judgment state (the thrust state (judgment)): All placement objects are placed in the placement area State

  Further, based on at least one of the plan creation stagnation time and the operation speed change value, the state of wonder in the user placement work is estimated. As for the state of ambiguity, it is considered that there is a selection target ambiguity as to which target object is arranged in a certain arrangement area, and an installation place ambiguity as to which arrangement area a certain object is arranged. . The selection target hesitation can be estimated based on the plan creation stagnation time, and the installation place hesitation can be estimated based on the operation speed change value. These states of hesitation are expressed as a degree of hesitation depending on the presence or absence of the user's hesitation.

  Thereafter, the planning support apparatus 100 determines a proposal mode of a placement plan that is considered to be beneficial in the current placement work state based on the user situation (suggest creation function). The proposal mode is composed of a suggested distance that indicates whether the alternative proposed to the user is close to or far from the user placement plan, and a replacement pattern of the objects placed in the placement area in the user placement plan. It is determined based on the progress of the placement work and the degree of hesitation. The planning support apparatus 100 creates a target alternative for the placement area based on the determined proposal mode and presents it to the user.

[1-3. GUI example]
Here, FIG. 3 to FIG. 5 show an example of an input screen 200 on which the user inputs an object arrangement plan for the arrangement area in the planning support apparatus 100, and FIG. 6 shows an alternative plan to the user by the planning support apparatus 100. An example of the proposal screen 300 to be displayed is shown. 3 to 6, the object placement operation for determining the arrangement of the display objects (corresponding to the placement objects) in the museum will be taken as an example. The layout of the display objects in the museum in this example is composed of six display groups in which the objects are grouped three by three. At this time, the user decides the arrangement of the object in consideration of the constraint condition where the arrangement place of the object is limited, the overall sense of unity, the amount of customers attracted by the exhibition object, and the like.

  The planning support apparatus 100 displays an input screen 200 as shown in FIG. 3 on the display device, for example. The input screen 200 includes an arrangement area 210 in which a layout represented by a plurality of arrangement areas 212 is displayed, and an object 222 not arranged in the arrangement area 212 among the plurality of objects 222 arranged in the arrangement area 212. A displayed object area 220 is provided. As shown in FIG. 4, the user operates a pointer 202 which is an information operation unit on the input screen 200 to select one target object 222 from the target object area 220 and drag and drop the selected target object 222. The desired placement area 212 is moved by a drop operation. The pointer 202 is operated based on input information from an operation input device such as a mouse or a keyboard. The user repeats this operation to arrange the object 222 in the arrangement area 212 of the arrangement area 210 and create an arrangement plan. Finally, a user arrangement plan in which the objects 222 are arranged in all the arrangement areas 212 as shown in FIG. 5 is completed.

  As shown in FIG. 3, the input screen 200 can also be provided with a selection target information display area 230 for displaying information related to the currently selected target 222. Accordingly, the user can place the object 222 in the placement area 212 in consideration of detailed information regarding the object 222. Further, an elapsed time display area 240 for displaying the elapsed time since the user started the input screen 200 may be provided on the input screen 200.

  When the determination button 250 is pressed by the pointer 202 in a state where the objects 222 are arranged in all the arrangement areas 212 on the input screen 200, the arrangement plan currently displayed on the input screen 200 is adopted. Further, when the proposal button 260 is pressed by the pointer 202 on the input screen 200, a request for an alternative proposal is output to the planning support apparatus 100. The planning support apparatus 100 that has received an alternative proposal request creates an alternative with the function shown in FIG. 2, and presents the alternative to the user on a proposal screen 300 as shown in FIG. 6, for example.

  In the proposal screen 300, an arrangement area 310 similar to the arrangement area 210 displayed on the input screen 200 is displayed. The arrangement area 312 of the arrangement area 310 is an alternative created by the planning support apparatus 100. An object 222 is arranged based on the plan. The user refers to the alternative displayed on the proposal screen 300 and presses the OK button 320 using the pointer 202 when adopting the alternative. The adopted alternative is reflected in the arrangement area 210 of the input screen 200. On the other hand, when the alternative from the planning support apparatus 100 is not adopted, the user presses the cancel button 330 with the pointer 202. When the cancel button 330 is pressed, the input screen 200 before the proposal screen 300 is displayed is displayed on the display device.

  The proposal screen 300 includes a questionnaire area 340 such as a similarity questionnaire 342 for examining the similarity between the alternative plan by the planning support apparatus 100 and the user placement plan, and a contribution questionnaire 344 for examining the contribution degree of the alternative to the user. May be provided. The questionnaire results acquired in the questionnaire area 340 can be used for verification of the effectiveness of the planning support apparatus 100, and can be reflected in the accuracy of the alternatives created by the planning support apparatus 100.

  In addition, the display device that displays the input screen 200 and the proposal screen 300 and the operation input device on which the user performs operation input may be a separate device from the planning support device 100, and is formed integrally with the planning support device 100. It may be. Further, the display device and the operation input device can be integrally formed by using a touch panel or the like, and thereby the user can intuitively input the operation with a finger or the like.

  As described above, the user can create an arrangement plan of the object with respect to the arrangement area, and can receive an alternative plan from the planning support apparatus 100. Hereinafter, creation of an alternative presented by the planning support apparatus 100 will be described in more detail.

<2. Create alternatives with planning support device>
[2-1. Functional configuration of planning support device]
First, based on FIGS. 7-11, the function structure of the planning assistance apparatus 100 which concerns on this embodiment is demonstrated. FIG. 7 is a functional block diagram showing a functional configuration of the planning support apparatus 100 according to the present embodiment. FIG. 8 is an explanatory diagram showing the relationship between the planned filling rate and the progress of the placement work. FIG. 9 is an explanatory diagram illustrating an example of a user situation pattern. FIG. 10 is an explanatory diagram illustrating an example of the proposal mode. FIG. 11 is an explanatory diagram illustrating an example of alternative adjustment information.

  As shown in FIG. 7, the planning support apparatus 100 according to the present embodiment includes an interface 110, an operation history acquisition unit 120, a modeling unit 130, a proposal mode determination unit 140, an alternative creation unit 150, and a timer unit. 160, an operation history storage unit 170, and an alternative adjustment information storage unit 180.

  The interface 110 is a functional unit that receives an operation input using a GUI from a user and an evaluation input for an alternative from an operation input device such as a mouse or a touch panel. The interface 110 outputs the input information input by the operation input device to the operation history acquisition unit 120. The interface 110 outputs the alternative created by the planning support apparatus 100 to the display device. By displaying the alternative on the display device, the alternative can be presented to the user. Here, the planning support device 100 has been described as a separate device from the operation input device and the display device on which the user performs operation input, but the present invention is not limited to such an example, and the planning support device 100 includes: It is also possible to provide an operation input device and a display device on which a user performs operation input.

  The operation history acquisition unit 120 acquires information related to the user's operation input acquired via the interface 110, and calculates information necessary for estimating a user situation by the modeling unit 130 described later. The operation history acquisition unit 120 acquires, as information related to a user's operation input, an arrangement state of an object on the GUI and a moving speed of an operation unit for moving the object or pressing a button on the GUI. To do. The operation means is, for example, a mouse pointer 202 as shown in FIG. 3, a finger that touches the GUI screen when the operation input device is configured by a touch panel, a dedicated stylus pen, or the like. The operation history acquisition unit 120 acquires the position information and time of the operation means in the GUI coordinate system, and uses the filling rate calculation unit 122, the stagnation time calculation unit 124, and the operation speed calculation unit 126 to perform the planned filling rate and plan creation stagnation. Calculate time and operating speed.

  The filling rate calculation unit 122 calculates a planned filling rate that represents the ratio of the arrangement area in which the objects are arranged among all the arrangement areas. The planned filling rate takes a value of 1 to 100 [%]. For example, as shown in FIG. 3, the planned filling rate when no object is placed in any placement area is 0%, and 8 placement areas out of 18 placement areas as shown in FIG. The planned filling rate when an object is placed in the box is about 44% (= 8/18 × 100). And as shown in FIG. 5, the plan filling rate in case the target object is arrange | positioned at all the arrangement | positioning area | regions will be 100%.

  The stagnation time calculation unit 124 calculates an elapsed time (plan creation stagnation time) from the start of the execution of the final event by the operation means to the present. The final event is the last operation performed retroactively from the time when the operation is performed. As the final event executed by the operation means, for example, the operation in which the object is finally arranged in the arrangement area on the GUI And an operation of selecting an object and an operation of closing a proposal screen 300 displaying an alternative by the planning support apparatus 100. The stagnation time calculation unit 124 uses the timer unit 160 to acquire the time from when such an operation is performed until the next operation is performed. When a new operation is performed during the elapsed time count, the stagnation time calculation unit 124 resets the elapsed time count by the timer unit 160 and starts counting the elapsed time since the new operation was performed.

The operation speed calculation unit 126 calculates the movement speed of the operation means based on the position information and time of the operation means in the GUI coordinate system. Then, the operation speed calculation unit 126 detects how the movement speed of the operation means has changed with respect to the average value within a preset time based on the movement speed of the operation means. Specifically, the operation speed calculating unit 126, for example, counts the number of times the moving speed of the operating means exceeds the average value and the number of times the operation speed change value n _v.

  For example, when the moving speed of the operating means changes from “below average → over average → below average”, the moving speed of the operating means exceeds the average value once, so the operating speed change value is 1 Further, in the case of a change of “below average → over average → under average → over average → under average”, the number of times the movement speed of the operating means exceeds the average value is 2, so the operation speed change value is 2. It becomes. Here, the operation speed change value is the number of times that the movement speed of the operation means exceeds the average value, but the present invention is not limited to such an example, and for example, the movement of the operation means within a preset time. It may be the number of times the speed is less than the average.

  The operation history acquisition unit 120 outputs the calculated planned filling rate, plan creation stagnation time, and operation speed change value to the modeling unit 130 and records them in the operation history storage unit 170.

  The modeling unit 130 estimates the situation of the user's placement work based on the planned filling rate, the plan creation stagnation time, and the operation speed change value calculated by the operation history acquisition unit 120, and represents the situation as a frequency. The modeling unit 130 includes a progress degree estimating unit 132 that estimates the progress of the user's placement work, and a doubt level estimation unit 134 that estimates the state of the doubt in the user's placement work.

The progress estimation unit 132 estimates the progress of the user's placement work based on the planned filling rate. As described with reference to FIG. 2, in the present embodiment, the progress of the placement work includes the placement object browsing state, the first draft creation state, the thrust divergence state (the thrust state (divergence)), the thrust convergence state (the thrust state ( Convergence)), and a thrust determination state (thrust state (determination)). The progress of the placement work will be described with reference to FIG. 8. First, the placement object browsing state is a state in which no object is placed in the placement area, that is, a plan filling rate is 0%. It can be said that the user is in the stage of browsing the arrangement object and examining the arrangement plan. The frequency in this state is P (x ₁ ). Here, x ₁ is used as a symbol to mean placement object access state, P (x ₁₎ is a numerical value that represents the certainty of being the arranged object browsing state. The frequencies P (x ₂ ) to P (x ₅ ) in the states described below also represent the likelihood of being in that state, similar to the frequencies P (x ₁ ).

Next, the first draft creation state is a state from when an object starts to be arranged in the arrangement area until the object is arranged in all the arrangement areas, that is, in a state where the planned filling rate is greater than 0% and less than 100%. is there. As shown in FIG. 8, as the time passes, the object is gradually arranged in the arrangement region, and the frequency in this state is P (x ₂ ). The state in which the objects are arranged in all the arrangement areas for the first time from the first draft creation state is the first draft creation completion state. Note that the first plan creation completion state is one of the drafting determination states.

  The state after the first plan creation completion state and until it is determined as one placement plan is the drafting state. The pushing state is a state after the objects are once arranged in all arrangement regions, that is, a state after the planned filling rate reaches 100% once. As shown in FIG. 8, the thrust state is further divided into three states: a thrust divergence state (a thrust state (divergence)), a thrust convergence state (a thrust state (convergence)), and a thrust judgment state (a thrust state (determination)). Can be classified.

A state in which objects are arranged in all the arrangement regions, that is, a state in which the planned filling rate is 100% is set as a pushing judgment state (pushing state (judgment)). The frequency in this state is P (x ₃ ). In addition, if the planned filling rate decreases after the planned filling rate reaches 100% once, it is considered that the object once placed is removed from the placement region and reconsidered. State (pushing state (divergence)). The frequency of the thrust divergence state is P (x ₄ ). Then, after the planned filling rate reaches 100% once, the planned filling rate decreases, and then when the planned filling rate increases again, the object once arranged is rearranged in the arrangement region. This state is assumed to be a thrust convergence state (thrust state (convergence)). The frequency of the thrust convergence state is assumed to be P (x ₅ ).

The frequencies P (x ₁ ) to P (x ₅ ) of each state may take either an integer value of 0 or 1, and a real number (0 ≦ P (x _i ) ≦) representing an intermediate frequency. 1, i = 1 to 5). The progress estimation unit 132 outputs the frequency of each state as the progress.

  The hesitation degree estimation unit 134 estimates the hesitation state in the user's placement work based on the plan creation stagnation time t and the operation speed change value. As described with reference to FIG. 2, in this embodiment, as a user's hesitation state, a selection target that is wondering which object is to be placed in a certain placement area, and which placement area is the target object. It is estimated that the installation location is in doubt about whether to place it. Therefore, the ambiguity estimation unit 134 according to the present embodiment includes a selection target ambiguity estimation unit 135 and an installation location ambiguity estimation unit 136.

The selection target stray degree estimation unit 135 estimates a selection target stray based on the plan creation stagnation time t. The selection target hesitation is represented by a selection target hesitation level P (y ₁ ) that is the frequency of the user's hesitation state and a selection target hesitation level P (y ₂ ) that is the frequency of the user's hesitation state. Incidentally, y ₁ is the symbol that represents the state where there is doubt on the selection, y ₂ is a symbol representing a state where no hesitation in selecting the object. Here, it has been found from preliminary experiments that the frequency P (y ₁ ) of the state in which the selection target is in doubt has a correlation with the plan creation stagnation time t. Generally, if the plan creation stagnation time t is long, the selection target There is a tendency to get lost. Accordingly, it is appropriate to express the selection target frequency P (y ₁ ) using the sigmoid function shown in the following formula (1).

Here, a ₁ and a ₂ are constants that are adjusted and set so that the result of the preliminary experiment and the value of Equation (1) are best matched. As the goodness of fit between the preliminary experiment result and the value of equation (1), for example, these square errors can be used, and the coefficient a in equation (1) is such that the square error with the preliminary experiment result is minimized. _1, to determine the value of _{a 2.} Further, since there are only two states of “selection confused” and “no confused”, the selection target confused frequency P (y ₂ ) is calculated from the selection target confused frequency P (y ₁ ). be able to. For example, if the frequency P (y ₁ ) with a selection target is a probability value, the frequency P (y ₂ ) = 1−P (y ₁ ) without a selection target is obtained. The selection target hesitation level estimation unit 135 outputs the selection target hesitation frequency P (y ₁ ) and the selection target hesitation absence frequency P (y ₂ ) as the selection target hesitation level.

  Note that the selection target hesitation level estimation unit 135 can also estimate the user's selection target hesitation based on the movement of the operation means before the target selection instead of the plan creation stagnation time t. For example, when the pointer 202 passes over the predetermined object a plurality of times before the object is selected by the pointer 202, or the pointer 202 moves in the vicinity of the predetermined object. It is considered that there is a low possibility that the selection target is lost. On the other hand, when the pointer 202 passes all over the objects, it is considered that there is a high possibility that the selection target is lost. The movement of the operation means can be estimated from the position information history of the operation means, the direction of the operation speed, and the like.

The installation location ambiguity estimation unit 136 estimates installation location ambiguity based on the operation speed change value. The installation location ambiguity is expressed by an installation location ambiguity P (z ₁ ) in a state where there is a user's ambiguity and an installation location ambiguity P (z ₂ ) in a state where there is no user's ambiguity. Note that z ₁ is a symbol representing a state where the installation location is unclear, and z ₂ is a symbol representing a state where the installation location is unclear. Here, power P in the state where there is doubt the location (z ₁₎ is correlated with the operating speed change value n _v, generally lost the location of the object selected as the operation velocity change value n _v is larger Tend to be. From this, it is appropriate to represent the frequency P (z ₁ ) at which the installation place is lost using the sigmoid function shown in the following mathematical formula (2), as in the case of the selection target.

Here, a ₃ and a ₄ are constants that are adjusted and set so that the result of the preliminary experiment and the value of Equation (2) are best matched. As the goodness of fit between the preliminary experiment result and the value of Equation (2), for example, these square errors can be used, and the coefficient a in Equation (2) is such that the square error with the preliminary experiment result is minimized. _3, to determine the value of _{a 4.} In addition, since there are only two states, “there is no hesitation” and “no hesitation”, the frequency P (z ₂ ) of no installation location is calculated from the frequency P (z ₁ ) of the installation location. be able to. For example, if the frequency P (z ₁ ) with the installation place is in doubt, the frequency P (z ₂ ) = 1−P (z ₁ ) without the installation place is obtained. The installation place hesitation degree estimation unit 136 outputs the installation place hesitation frequency P (z ₁ ) and the installation place no hesitation frequency P (z ₂ ) as the installation place hesitation degree.

  In addition, the installation place hesitation degree estimation unit 136 can also estimate the installation place hesitation based on the operation speed of the operation means instead of the operation speed change value. For example, when the object 202 is selected and the pointer 202 passes over a plurality of placement areas, it is highly likely that the placement location is lost. Also, if the direction of movement of the operating means has changed several times based on the operating speed, it is estimated that the operating means continues to move, so it is highly likely that the user has lost the location. It is done.

In the present embodiment, the sigmoid function is used to represent the relationship between the selection target ambiguity and the plan creation stagnation time, and the relationship between the installation location ambiguity and the operation speed change value, but the present invention is not limited to such an example. These relationships may be expressed in other functional forms. For example, for the installation location ambiguity, the degree of ambiguity when the operation speed change value is n is calculated from a preliminary experiment, and an array of the operation speed change value _nv and the ambiguity is generated in advance, and the operation speed is based on the arrangement. the change value n _v may be calculated to stray degree corresponding.

  When the modeling unit 130 estimates the progress of the user's placement work and the state of wonder in the user's placement work, the modeling unit 130 calculates the user situation in the placement work based on the estimation result. The user situation is represented by 20 patterns shown in FIG. In FIG. 9, all combinations of the progress of the placement work, the selection object doubt, and the installation place doubt are represented as user situation patterns. The modeling unit 130 calculates the frequency of each current user situation pattern and outputs it to the proposal mode determination unit 140. Here, the frequency of the user situation pattern is a numerical value indicating the probability of each user situation pattern. When the numerical value is a probability, the total value of the frequencies of all the user situation patterns is 1 (= 100%). ) For example, when the frequency of the user situation pattern 7 of FIG. 9 is calculated as 0.8, the frequency of the user situation pattern 8 is 0.2, and the frequency of the other user situation patterns is calculated as zero, the user is in the first draft creation state. It means an estimation result that the installation location is lost with a probability of 80%.

  The proposal mode determination unit 140 determines an alternative proposal mode to be presented to the user based on the frequency of the user situation pattern. As described with reference to FIG. 2, the proposal mode of the alternative by the proposal mode determination unit 140 includes a proposal distance indicating whether the alternative proposed to the user has contents close to or far from the user placement proposal, and the user The arrangement plan includes an object replacement pattern arranged in the arrangement area.

The proposed distance is compared to the previous proposed alternative, whether an alternative closer to the user placement plan is proposed (proposed distance: nearer mode), a far alternative is proposed (proposed distance: farther mode), or the previous There are three modes: suggesting an alternative with the same degree of similarity as the alternative (proposed distance: middle mode). The calculation of the proposed distance is performed by repeatedly exchanging a pair of objects with respect to the user's arrangement plan, performing an operation for matching the alternative plan by the planning support apparatus 100, and obtaining a case where the number of exchanges is minimized. . At this time, the weighting of the exchange may be changed depending on whether the object exchange belongs to the same group or a different group. Thus, the distance d and the similarity s of the placement plan are expressed by the following equation using the number of exchanges c ₁ and its weight w ₁ within the same group, and the number of exchanges c ₂ and the weight w ₂ between different groups. It is represented by (3) and formula (4). Note that d _max in Equation (4) is the maximum value of distance, and is the value of d when all exchanges must be performed between different groups.

  The relationship between the similarity s and each mode of the proposed distance (nearer / middle / farther) is felt by the user with respect to the similarity of the alternative calculated by the planning support apparatus 100 obtained by the preliminary experiment and the alternative. A threshold value is set in advance based on the relationship with the distance. Then, when the proposal distance in the proposal mode is nearer, the planning support apparatus 100 presents an alternative that is closer to the user's placement plan than the previous time when the similarity is higher than the threshold value compared to the previous alternative. . In addition, when the proposed distance in the proposal mode is farther, the planning support apparatus 100 presents an alternative that is farther from the user's placement plan than the previous time, which is decreased by a threshold value or more. Then, when the proposed distance in the proposal mode is middle, the planning support apparatus 100 presents an alternative with a distance similar to the previous time in which the change in similarity is within a threshold value.

  The replacement pattern of the objects arranged in the arrangement area represents whether or not the arrangement of some of the objects in the user's arrangement plan is fixed and presented in the alternative. When the user is performing the placement work, useful information differs depending on whether the object is hardly placed in the placement area or if the object is placed in all the placement areas. For example, when almost no object is placed in the placement area, a placement plan that is completely different from the user's placement plan is helpful for the user. On the other hand, after the objects have been placed in all the placement areas, even if a placement plan that is completely different from the user's placement plan is proposed, it is not very helpful for the user, but rather a fine correction plan that is close to the user's placement plan is presented. This is useful information for the user.

  Therefore, the planning support apparatus 100 creates an alternative that fixes a part of an object that is often selected by the user according to the user situation (fixed upper item), and is one of the objects that are hardly selected by the user. It is determined whether an alternative is to be presented with a replacement pattern of creating an alternative with a fixed part (fixing subordinate items) or creating an alternative without fixing a specific object (no fixing).

  As an intention to fix the upper item with high operation frequency, for example, in the first draft creation state with a low planned filling rate, the object placed in the arrangement area becomes the upper item with high operation frequency, and the unplaced object Is a sub-item with a low operation frequency. At this time, if a higher-order item with high operation frequency is moved, the arrangement of the target object for which the user has determined the arrangement destination is changed, which may confuse the user. In such a case, it is considered effective to fix the upper item having a high operation frequency.

  On the other hand, as an intention to fix a lower-level item with low operation frequency, for example, an object that is hardly selected after being placed in the placement area in the selection state may have been determined by the user to be appropriately placed. Is expensive. At this time, if a lower-level item with low operation frequency is moved, the arrangement of the object for which the user has determined the arrangement destination is changed, which may confuse the user. In such a case, it is considered effective to fix the upper item whose operation frequency is low.

  In addition, as an intention not to fix the target object, for example, when the target object to be arranged in the arrangement area is selected with the same frequency in the pushing state, the user cannot determine the arrangement plan. It may have fallen. In such a case, a new viewpoint can be shown by making a significant change to the user's arrangement plan, and there is a high possibility that an effective alternative can be presented. Note that the number of objects to be fixed can be set as appropriate. For example, among all the objects, the upper or lower third item of the selection frequency may be fixed.

  The proposal mode determination unit 140 determines an alternative proposal mode based on the current user situation. The proposal mode is represented by all combinations of the proposal distance and the replacement pattern, and is represented by nine proposal modes a to i shown in FIG. The proposal mode determination unit 140 refers to the alternative adjustment information described later, acquires one or a plurality of proposal modes that may be selected, and the degree of selectivity thereof. One proposal mode is determined. The alternative adjustment information is information stored in advance in an alternative adjustment information storage unit 180, which will be described later. For example, the relationship between the user situation pattern and the proposal mode as shown in FIG. 11 is represented in a two-dimensional array. It is. More specifically, the alternative adjustment information represents the frequency of the proposal mode in FIG. 10 on the condition of the user situation pattern shown in FIG. 9, and this frequency is selected by each proposal mode in each user situation pattern. Represents the ratio (selectivity). When the frequency is a probability, the two-dimensional array shown in FIG. 11 becomes a conditional probability distribution table.

  For example, in the case of the user situation pattern 1, the proposal mode a is selected at a frequency of 0.20, the proposal mode b is selected at a frequency of 0.35,..., And the proposal mode i is selected at a frequency of 0.13. It is considered that the proposal mode having a high selected ratio is highly likely to be an effective proposal in the user situation.

In determining one alternative proposal mode, the proposal mode determination unit 140 first determines the frequency of each state calculated by the modeling unit 130 based on the alternative adjustment information as shown in FIG. The degree of selectivity of each proposal mode is calculated based on the degree of progress and the level of hesitation. Specifically, for example, the frequency of the user situation pattern 1 is P based on the arrangement object browsing state frequency P (x ₁ ), the selection target suspicious frequency P (y ₁ ), and the installation location suspicious frequency P (z ₁ ). (X ₁ ) P (y ₁ ) P (z ₁ ). Also, the frequency of the user situation pattern 2 is disposed object browsing state power P _(x 1), selection hesitation Yes power P _(y 1), from the location hesitation without power _{P (z 2), P (} x 1) P (y ₁ ) P (z ₂ ). Similarly, the frequency of the user situation pattern 3 is P (x ₁ ) P (y ₂ ) P (z ₁ ), and the frequency of the user situation pattern 4 is P (x ₁ ) P (y ₂ ) P (z ₂ ). .

In this way, the frequencies of all user situation patterns are calculated. Accordingly, the selectivity P (a) of the proposal mode a row is calculated by the following equation (5). Here, P (x _i ) P (y _j ) P (z _k ) is the frequency of the user situation pattern, and P (a | x _i , y _j , z _k ) is the user situation in the proposed mode a column in FIG. It is a value of a two-dimensional array corresponding to a pattern.

  The proposal mode determination unit 140 selects one proposal mode for creating an alternative based on the selectivity of each proposal mode in the current user situation. For example, in the current user situation, it is assumed that the selectivity of the proposal mode a is 0.8, the selectivity of the proposal mode b is 0.2, and the selectivity of the other proposal modes is 0. At this time, the alternative creation unit 150 determines one proposal mode using a random number so that the proposal mode a is selected with a frequency of 80% and the proposal mode b with a frequency of 20%. Or the proposal mode determination part 140 may determine the proposal mode with the highest selectivity as a proposal mode which produces an alternative. Then, the proposal mode determination unit 140 outputs the determined proposal mode to the alternative creation unit 150.

  The alternative creation unit 150 creates an alternative based on the proposal mode determined by the proposal mode determination unit 140. For example, when proposing an alternative in the proposal mode a, the upper item is fixed, and an alternative that is farther from the user's arrangement plan than the previously presented alternative is calculated. The alternative creation unit 150 presents the alternative created through the interface 110 to the user.

  The timer unit 160 measures time according to an instruction from a functional unit constituting the planning support apparatus 100. For example, the timer unit 160 counts the time from the start of the last event execution by the operation means based on the instruction of the stagnation time calculation unit 124 and notifies the stagnation time calculation unit 124 of the time.

  The operation history storage unit 170 is a storage unit that stores an operation history acquired via the interface 110. The operation history storage unit 170 stores the position information of the operation means on the GUI in association with the time at that time. The operation history storage unit 170 can also store evaluations for alternatives by the planning support apparatus 100.

  The alternative plan adjustment information storage unit 180 is a storage unit that stores alternative plan adjustment information used when the proposal mode determination unit 140 determines a proposal mode of an alternative plan. The alternative adjustment information is information representing the selectivity of each proposal mode with respect to the user situation pattern in a two-dimensional array, and is a conditional probability distribution table when the selectivity is a probability value. The alternative adjustment information is information acquired in advance by conducting a preliminary experiment. In this embodiment, the selectivity of each proposal mode with respect to the user situation pattern is acquired by learning based on, for example, Bayesian estimation (see Non-Patent Document 6 above).

  For example, when the selectivity of the proposal mode is a probability value, first, the initial values of the selectivity are all set to the same value (1 / total number of proposed modes), and an alternative according to the selectivity is presented to the user. Then, when the user evaluates the proposed alternative, and is evaluated as a “good” proposal, the selectivity (probability value) of the proposal mode is increased, and the other selectivity (probability value) is relatively decreased. To do. On the other hand, if the user evaluates the proposed alternative as a “bad” proposal, the selectivity (probability value) of the proposal mode is reduced and the other selectivity (probability value) is relatively increased. . Such adjustment of selectivity can be easily performed using a Bayesian network construction support system. The selectivity of each proposal mode for the user situation pattern finally obtained in this manner is recorded in the alternative plan adjustment information storage unit 180 as alternative plan adjustment information.

[2-2. Alternative creation process by planning support device]
The alternative creation process by the planning support apparatus 100 shown in FIG. 7 will be described with reference to FIGS. 12 and 13. FIG. 12 is a flowchart showing an alternative creation process by the planning support apparatus 100 according to the present embodiment. FIG. 13 is a flowchart showing an alternative creation process based on the determined proposal mode.

  In the alternative plan creation process by the planning support apparatus 100 according to the present embodiment, as shown in FIG. 12, the operation history acquisition unit 120 first acquires the operation history of the GUI by the user via the interface 110 (S100). . The operation history acquisition unit 120 acquires the position information and time of the operation means in the GUI coordinate system, and uses the filling rate calculation unit 122, the stagnation time calculation unit 124, and the operation speed calculation unit 126 to perform the planned filling rate and plan creation stagnation. Time and operation speed are calculated, and an operation speed change value is further calculated from the operation speed.

  Next, the progress degree estimation unit 132 calculates the progress degree based on the planned filling rate (S110), and the lost degree estimation unit 134 calculates the degree of doubt (S120). In step S120, at least one of the selection target ambiguity and the installation location ambiguity is calculated. In steps S110 and S120, either process may be executed first, or these processes may be executed in parallel. When the degree of progress and the degree of hesitation are calculated in steps S110 and S120, the modeling unit 130 can calculate the frequency of each current user situation pattern (S130).

  When each user situation pattern is calculated by the modeling unit 130, the proposal mode determination unit 140 determines one alternative proposal mode (S140). The proposal mode unit 140 determines an alternative proposal mode according to the selectivity of each proposal mode in the current user situation pattern based on the alternative adjustment information. Thereafter, the alternative plan creation unit 150 creates an alternative plan based on the proposal mode determined by the proposal mode determination unit 140 (S150). The alternative creation unit 150 creates an alternative by the process shown in FIG. 13 based on a preset evaluation function that indicates the quality of the target object placement result.

  First, the alternative creation unit 150 randomly generates M kinds of solutions that satisfy the placement constraints including the presence / absence of fixation of the object based on the replacement pattern of the proposal mode, here, the placement results of the object with respect to the placement area (S151). ). In step S151, an object placement result is generated based on the suggestion mode replacement pattern determined in step S140 of FIG. That is, when the replacement pattern is fixed to the upper item, the arrangement result of the predetermined number of objects that are frequently selected by the user is not changed, and the arrangement of the other objects is randomly replaced, and the arrangement result satisfies the arrangement constraint. Only M is stored. Similarly, when the replacement pattern is fixed to the lower item, the arrangement of the predetermined number of objects with low user selection frequency is not changed, and the arrangement of the other objects is randomly changed to satisfy the arrangement constraint. Only M results are stored. When the replacement pattern is not fixed, all objects are replaced at random.

  Then, the alternative creation unit 150 calculates an evaluation value for each of the generated M arrangement results using an evaluation function, and leaves the arrangement result with the best evaluation value (S152).

  Next, the alternative creation unit 150 replaces the two sets of objects among the objects that form the arrangement result of the best solution left in step S152, and generates M new solutions (S153). An evaluation value is calculated using an evaluation function for (S154). Then, the alternative creation unit 150 determines whether or not there is a solution with a higher evaluation value than the current best solution among the new solutions (S155), and there is no solution with a higher evaluation value than the current best solution. In step S157, the current best solution is determined as an alternative.

  On the other hand, when there is a solution having a higher evaluation value than the current best solution in step S155, the alternative creation unit 150 determines whether or not the number n of repetitions of the process from step S153 in FIG. Determination is made (S156). When the number of repetitions n is the predetermined number N in step S156, the process proceeds to step S157, and the current best solution is determined as an alternative). On the other hand, when the number of repetitions n is not the predetermined number N (n <N) in step S156, a new solution having a higher evaluation value than the current best solution is determined as the best solution (S158). Then, after adding 1 to the number of repetitions n of the process from step S153 (S159), the process from step S153 is repeated.

  Thus, the alternative creation unit 150 creates an alternative presented to the user. The algorithm in FIG. 13 uses a simple hill-climbing method that replaces two sets of objects as the arrangement result, but the present invention may use other algorithms. Further, the number M for generating the arrangement result of the object with respect to the arrangement area and the number N of repetitions of steps S153 to S156 are constants set by preliminary experiments. For example, if the number of arrangement areas is about 18, it is appropriate to set values of about M = 10 and about N = 30. In addition, when the number of arrangement | positioning area | regions or the number of objects is more than 18, it is good to make M and N relatively large.

  Heretofore, the configuration of the planning support apparatus 100 according to the present embodiment and the alternative creation process by this have been described. According to the planning support apparatus 100 according to the present embodiment, the progress of the placement work and the user's lost state are estimated based on the user's operation history. Then, the planning support apparatus 100 estimates an appropriate proposal mode according to the estimated user situation, and creates an alternative in the proposal mode. This makes it possible to present an effective alternative according to the situation of the person in charge (the user of the device) who performs the arrangement work, thereby reducing the burden on the person in charge. Further, the planning support apparatus 100 may be able to propose an alternative that the person in charge did not notice, and a better arrangement result can be obtained. Further, when an alternative similar to the arrangement result of the person in charge is proposed from the apparatus, the confidence in the created arrangement result is increased and unnecessary trial and error is eliminated, so that the planning time is shortened.

<3. Hardware configuration example>
Next, the hardware configuration of the planning support apparatus 100 according to the embodiment of the present invention will be described in detail with reference to FIG. FIG. 14 is a block diagram illustrating a hardware configuration example of the planning support apparatus 100 according to the embodiment of the present invention.

  The planning support apparatus 100 mainly includes a CPU 901, a ROM 903, and a RAM 905. The planning support apparatus 100 further includes a bus 907, an input device 909, an output device 911, a storage device 913, a drive 915, a connection port 917, and a communication device 919.

  The CPU 901 functions as an arithmetic processing unit and a control unit, and controls all or a part of the operation in the planning support apparatus 100 according to various programs recorded in the ROM 903, the RAM 905, the storage apparatus 913, or the removable recording medium 921. The ROM 903 stores programs used by the CPU 901, calculation parameters, and the like. The RAM 905 primarily stores programs used by the CPU 901, parameters that change as appropriate during execution of the programs, and the like. These are connected to each other by a bus 907 constituted by an internal bus such as a CPU bus.

  The bus 907 is connected to an external bus such as a PCI (Peripheral Component Interconnect / Interface) bus via a bridge.

  The input device 909 is an operation unit operated by the user, such as a mouse, a keyboard, a touch panel, a button, a switch, and a lever. The input device 909 may be, for example, remote control means (so-called remote control) using infrared rays or other radio waves, or may be an external connection device 923 such as a PDA corresponding to the operation of the planning support device 100. May be. Furthermore, the input device 909 includes, for example, an input control circuit that generates an input signal based on information input by a user using the operation unit and outputs the input signal to the CPU 901. The user of the planning support apparatus 100 can input various data and instruct a processing operation to the planning support apparatus 100 by operating the input device 909.

  The output device 911 is configured by a device that can notify the user of the acquired information visually or audibly. Examples of such devices include CRT display devices, liquid crystal display devices, plasma display devices, EL display devices and display devices such as lamps, audio output devices such as speakers and headphones, printer devices, mobile phones, and facsimiles. The output device 911 outputs results obtained by various processes performed by the planning support apparatus 100, for example. Specifically, the display device displays the results obtained by the various processes performed by the planning support device 100 as text or an image. On the other hand, the audio output device converts an audio signal composed of reproduced audio data, acoustic data, and the like into an analog signal and outputs the analog signal.

  The storage device 913 is a data storage device configured as an example of a storage unit of the planning support device 100. The storage device 913 includes, for example, a magnetic storage device such as an HDD (Hard Disk Drive), a semiconductor storage device, an optical storage device, or a magneto-optical storage device. The storage device 913 stores programs executed by the CPU 901, various data, various data acquired from the outside, and the like.

  The drive 915 is a recording medium reader / writer, and is built in or externally attached to the planning support apparatus 100. The drive 915 reads information recorded on a removable recording medium 921 such as a mounted magnetic disk, optical disk, magneto-optical disk, or semiconductor memory, and outputs the information to the RAM 905. The drive 915 can also write a record on a removable recording medium 921 such as a magnetic disk, an optical disk, a magneto-optical disk, or a semiconductor memory. The removable recording medium 921 is, for example, a CD medium, a DVD medium, a Blu-ray medium, or the like. The removable recording medium 921 may be a CompactFlash (registered trademark) (CompactFlash: CF), a flash memory, an SD memory card (Secure Digital memory card), or the like. Further, the removable recording medium 921 may be, for example, an IC card (Integrated Circuit card) on which a non-contact IC chip is mounted, an electronic device, or the like.

  The connection port 917 is a port for directly connecting a device to the planning support apparatus 100. Examples of the connection port 917 include a USB (Universal Serial Bus) port, an IEEE 1394 port, a SCSI (Small Computer System Interface) port, and an RS-232C port. By connecting the external connection device 923 to the connection port 917, the planning support apparatus 100 acquires various data directly from the external connection device 923 or provides various data to the external connection device 923.

  The communication device 919 is a communication interface including a communication device for connecting to the communication network 925, for example. The communication device 919 is, for example, a communication card for wired or wireless LAN (Local Area Network), Bluetooth (registered trademark), or WUSB (Wireless USB). The communication device 919 may be a router for optical communication, a router for ADSL (Asymmetric Digital Subscriber Line), or a modem for various communication. The communication device 919 can transmit and receive signals and the like according to a predetermined protocol such as TCP / IP, for example, with the Internet and other communication devices. The communication network 925 connected to the communication device 919 is configured by a wired or wireless network, and may be, for example, the Internet, a home LAN, infrared communication, radio wave communication, satellite communication, or the like. .

  Heretofore, an example of the hardware configuration capable of realizing the function of the planning support apparatus 100 according to the embodiment of the present invention has been shown. Each component described above may be configured using a general-purpose member, or may be configured by hardware specialized for the function of each component. Therefore, it is possible to change the hardware configuration to be used as appropriate according to the technical level at the time of carrying out this embodiment.

<4. Example>
Effects expected by using the planning support apparatus 100 according to the embodiment include an increase in the degree of confidence in the user's final decision plan and an increase in the degree of contribution to the creation of the user's placement plan. Therefore, using the planning support apparatus 100 of the above embodiment, 18 subjects are randomly selected from the objects to be displayed shown in FIG. 15 and arranged on the display layout shown in FIG. The effect was verified. 15, the classification representing the category of the display object, the size of the display object, the time when the display object was born, the material constituting the display object, and the transportation cost for the installation of the display object Information on the weight of the exhibition objects, the expected amount of customers attracted by the exhibition, and the first restriction indicating whether the exhibition group is installed first, etc., and the user considers the information about these exhibition objects. Arrange the display objects in the arrangement area.

  At this time, when the alternative is presented to the user by the planning support apparatus 100 with a certain degree of selectivity (without considering the situation), the user situation is estimated according to the operation history of the user, and the selectivity of the proposal mode is changed. Two types of placement work were performed, depending on the case (with consideration for the situation). Then, after completing the object placement work, each user is asked a questionnaire, and the confidence level felt by the user with respect to the final plan decided and the contribution degree of the planning support apparatus 100 in creating the user placement plan are 1 to 9 points. I got a score in 9 stages until. The average of each score of the questionnaire results is shown in FIG.

  From FIG. 16, the case where the selectivity of the proposal mode is changed in consideration of the user situation is higher for the arrangement result created by the user than in the case where the user situation is not taken into consideration. Recognize. This may be because the degree of confidence in the final decision is increased by presenting an alternative plan close to the user's placement plan by the planning support apparatus 100 at the stage where the user is pushing the placement plan. In addition, regarding the degree of contribution of the planning support apparatus 100, it is found that the case where the selection degree of the proposal mode is changed in consideration of the user situation is higher than the case where the user situation is not taken into consideration.

  The preferred embodiments of the present invention have been described in detail above with reference to the accompanying drawings, but the present invention is not limited to such examples. It is obvious that a person having ordinary knowledge in the technical field to which the present invention pertains can come up with various changes or modifications within the scope of the technical idea described in the claims. Of course, it is understood that these also belong to the technical scope of the present invention.

DESCRIPTION OF SYMBOLS 100 Planning support apparatus 110 Interface 120 Operation history acquisition part 122 Filling rate calculation part 124 Stagnation time calculation part 126 Operation speed calculation part 130 Modeling part 132 Progress degree estimation part 134 Loss degree estimation part 135 Selection target doubt degree estimation part 136 Installation place doubt Degree estimation unit 140 Proposal mode determination unit 150 Alternative creation unit 160 Timer unit 170 Operation history storage unit 180 Alternative adjustment information storage unit

Claims (11)

A planning support apparatus that supports planning of the arrangement of a plurality of objects in a plurality of arrangement areas,
A progress estimation unit that estimates a progress indicating the progress of the placement work, based on a planned filling rate that represents a ratio of the objects placed with respect to the placement region;
Based on an operation history of a user who arranges the object with respect to the arrangement area, a degree of ambiguity estimation unit that estimates a degree of ambiguity representing a user's ambiguity when arranging the object in the arrangement area;
A proposal mode determination unit that determines a proposal mode that represents the feature of the arrangement plan of the object with respect to the arrangement area, based on the progress degree and the degree of ambiguity;
An alternative creation unit that creates an alternative of the placement plan based on the proposal mode;
A planning support apparatus comprising: The suggestion mode determination unit
Based on alternative adjustment information that is generated in advance based on the operation history and represents the selectivity that is the ratio of the proposal mode to the planning status calculated from the degree of progress and the degree of hesitation, Calculating the degree of selectivity of the proposal mode corresponding to the degree of progress and the degree of hesitation,
The planning support apparatus according to claim 1, wherein one proposal mode is determined from a plurality of the proposal modes according to the degree of selectivity. The alternative plan adjustment information is information representing the selectivity of the proposal mode with respect to the user's planning status calculated from the progress degree and the degree of hesitation with a conditional probability, and is obtained using Bayesian estimation. The planning support apparatus according to claim 2, wherein the planning support apparatus is characterized.   The progress estimation unit is configured to display an arrangement object browsing state in which the object is not arranged with respect to the arrangement area based on the planned filling rate, and all the objects are arranged once with respect to the arrangement area. Representing the frequency of each state of the first draft creation state, which is a state before being placed, and the pushing state, which is a state after all the objects have been placed once with respect to the placement region, as the progress degree. The planning support apparatus according to any one of claims 1 to 3.   The progress estimation unit, as the frequency of the pushing state, the plan filling rate in which the plan filling rate decreases after the object is once placed in the placement region, and the plan filling rate from the thrust spreading state. The planning support apparatus according to claim 4, wherein the degree of progress is expressed as a degree of progress in a convergence state in which the object rises again and a state in which a determination is made in which all the objects are arranged.   6. The method according to claim 1, wherein the wonder level estimation unit estimates the frequency of a state of ambiguity and the frequency of a state without ambiguity representing a user's ambiguity state, and sets the degree of ambiguity. The planning support apparatus according to claim 1.   The plan according to claim 6, wherein the degree of ambiguity estimation unit estimates the degree of ambiguity based on a plan creation stagnation time that is an elapsed time from the last event for the object in the placement operation. Support device.   The planning support apparatus according to claim 6, wherein the hesitation degree estimation unit estimates the degree of hesitation based on an operation speed for operating the object in the placement work.   The ambiguity estimation unit estimates an object ambiguity representing the ambiguity of the object to be arranged with respect to the arrangement area, and an installation place ambiguity representing the ambiguity of the arrangement area to arrange the object. The planning support device according to any one of claims 6 to 8. A planning support method by a planning support apparatus that supports planning of the arrangement of a plurality of objects in a plurality of arrangement areas,
A progress degree estimating step for estimating a progress degree representing the progress of the placement work based on a planned filling rate representing a ratio of the object placed in the placement area by the progress degree estimation unit ;
A hesitation for estimating a hesitation level representing a user's hesitation state when placing the target object in the placement area based on a user's operation history of placing the target object in the placement area by a stray degree estimation unit A degree estimation step;
A suggestion mode determination unit for determining a proposal mode representing a feature of an arrangement plan of the object with respect to the arrangement area based on the progress degree and the degree of hesitation; and
An alternative creation step of creating an alternative of the placement plan based on the proposal mode by an alternative creation unit ;
A planning support method characterized by including: A computer program that causes a computer to function as a planning support device that supports planning of the arrangement of a plurality of objects in a plurality of arrangement areas,
A progress estimation unit that estimates a progress indicating the progress of the placement work, based on a planned filling rate that represents a ratio of the objects placed with respect to the placement region;
Based on an operation history of a user who arranges the object with respect to the arrangement area, a degree of ambiguity estimation unit that estimates a degree of ambiguity representing a user's ambiguity when arranging the object in the arrangement area;
A proposal mode determination unit that determines a proposal mode that represents the feature of the arrangement plan of the object with respect to the arrangement area, based on the progress degree and the degree of ambiguity;
An alternative creation unit that creates an alternative of the placement plan based on the proposal mode;
A computer program for causing a computer to function as a planning support apparatus.

Images