Matching Children to Daycare Centers While Best Meeting Applicants' Preferences Requires Extensive Labor and Time
In recent years, the number of children on waiting lists for daycare centers has become a social issue that the media has taken up. According to the Ministry of Health, Labour and Welfare (MHLW), the number of dual-income households has surpassed the number of households with full-time housewives in 1992. Then dual-income households have continued to increase year-over-year (*); however, daycare centers--primary places where working parents can leave their children--are full and have no vacancies, which is the cause of the waiting-list problem.
In addition to the waiting-list problem, matching children to daycare centers that fulfill the applicants' preferences is also a difficult task. Local governments repeatedly conduct trial and error every time, which requires much labor and time, to match children to daycare centers while trying to accommodate each family's preferences as much as possible, such as requests for siblings to be admitted to the same daycare center, prioritizing younger siblings over older ones or pulling out if only one child gets a place. In Saitama City, for example, the task of trying to match 7,959 children to 311 daycare centers while considering these complex requirements requires several days of work by 20 to 30 employees.
Putting their children in daycare centers is an urgent matter for dual-income households. Fujitsu and Kyushu University have jointly developed an AI-based matching technology to solve the problem of how to allocate children to a limited number of places fairly, quickly and meticulously.
- *: Changes in the Numbers of Households with a Full-Time Housewife and Dual-Income Households by the Ministry of Health, Labour and Welfare (MHLW)
Solving the Problem in Just Seconds Using Game Theory
Fujitsu and Kyushu University have now developed matching technology that can automatically determine the assignment pattern that will fulfill the preferences of as many applicants as possible, according to priority ranking, in consideration of complex requirements that are determined by human trial and error. This technology uses a mathematical approach called game theory, which rationally handles conflicts and cooperation among people in society where interests are not necessarily aligned.
As an example, one can consider assigning two sets of siblings (a total of four children) to two daycare centers (A and B) that can take two children each. There are six possible admission patterns (Figure 1). In this example, the parents of each child have requested daycare A over daycare B, but have also expressed that they would prefer that both siblings go to daycare B rather than be split up. In this situation, the rule is to fulfil these preferences as much as possible in determining admission assignment while also respecting the priority ranking of the children.
If the preferences for child 2, for example, cannot be met due to the preferences for child 1, who has higher priority, then that must be accepted, but if they cannot be met due to the preferences for child 3, who has lower priority, this would be a violation of the rule. In this way, it is necessary to check if the rule is being violated while considering both the priority of the children and their preferences. In addition, for siblings who have different priority rankings, where seat assignments 3 and 4 both fulfill the rules, assignment 3 is considered optimal, because it can meet the preferences of child 1, who has the highest priority ranking.
This is a simple example of assigning children to two daycare centers that can take two children each, but actually it is not easy to perform this selection process for thousands of applicants.
This technology was evaluated using anonymized data for about 8,000 children in Saitama City. The result was that this technology was able to calculate in just seconds seat assignments that fulfilled the complex and detailed requirements unique to Saitama City, which had previously taken 20 to 30 people several few days. This technology's accuracy is equivalent to manual admission screening in coordinating daycare facility usage, which can be considered as close to perfect as possible.
A Variety of Matching Applications from Deploying Personnel to Formulating Employee Work Schedules
When this technology is commercialized, assignment results can be obtained in seconds using AI, which could dramatically reduce the burden of local government employees assigning seats at daycare centers and enable the results to be sent to applicants earlier, thereby contributing to improving services for local residents. Knowing the results earlier also allows applicants to make plans for returning to their workplace well in advance.
By the end of fiscal 2017, Fujitsu plans to offer this technology as an optional service in its MICJET MISALIO Child-Rearing Solution, a childcare support system for local governments. In addition, as part of its Fujitsu Human Centric AI Zinrai artificial intelligence technology, Fujitsu is aiming to adopt this technology in a variety of matching applications beyond daycare admissions, such as fairly matching personnel deployments employee schedules within an organization.