What is IPSAC?

  • Integrated Planning for School and Community is a suite of tools and practices designed to assist decision makers in the school planning process.  IPSAC is a product of the Operations Research and Education Laboratory, located at the Institute for Transportation Research and Education on Centennial Campus at North Carolina State University.  IPSAC was a 1998 Edelman Prize Winner, awarded by the International Forum of Operations Research and Management Science (INFORMS), the world’s leading professional association of operations researchers and management scientists.  The planning system has been featured in numerous professional journals, newspapers, and media broadcasts.

What is OREd?

  • The Operations Research and Education Laboratory was founded in 1990 by Dr. Raymond Taylor (Professor Emeritus, College of Education, NCSU) to scientifically address politically sensitive school planning issues such as district membership projections, determining the location of new school sites, creating new attendance boundaries, and examining demographic balance.  OREd has served school districts ranging  from 4000 to 150,000 students in North Carolina, South Carolina, Virginia, and Mississippi.

How are school building capacities calculated?

  • Figuring out how many students a building can hold can be a complex and constantly changing calculation. Many North Carolina districts use a formula created by the NC Department of Public Instruction that combines student/teacher ratios, square footage, the number of special needs classrooms, and other factors.  These types of “architectural” formulas may not fully account for the varying classroom programs being used by a school.  OREd teams with a school architect or the district facilities expert to help answer this challenging question.

What is an “optimal” school site?

  • The “OR” in OREd stands for Operations Research, a branch of mathematics dealing with solving problems having many variables.  An optimal school site is one which minimizes student travel distance for all students in the impacted area while at the same satisfying constraints on building capacity or school demographics.  Optimal school sites generated by the IPSAC process are particularly useful to school districts that are experiencing growth by locating schools in areas best suited to accommodate for future growth.

What is a system-wide feeder system?

  • School systems usually have three grade levels:  an elementary school level (K-5), a middle school level (6-8), and a high school level (9-12).  A clean feeder system is one in which (a) the attendance zones at each level coincide, and (b) each elementary school that lies within a given middle school attendance zone sends all of its students to that middle school and all the middle schools within a high school attendance zone sends all of its students to that high school. There are obvious advantages to districts having a clean feeder system; however, it becomes difficult to maintain when growth is not uniform and when capacities at each level don’t match.  That is, for the schools in a clean feeder system to be fully utilized, the combined capacities of the middle schools which feed into a high school needs to be about 2/3 of that high school, and the combined capacity of the elementary schools that feed into a middle school needs to be a little more than twice than that of the middle school.

What is Out-of-District analysis?

  • Many systems have programs at one school that attract students outside of its base attendance zone. In addition, the district may allow students outside of a school’s base attendance zone to attend that school for a variety of reasons. Thus, an out-of-district analysis provides important information for the school and is a vital ingredient to decision-making with regard to attendance zone re-negotiations.

What are forecasting models?

  • The type of model used to create forecasts for school districts depends on several factors:
    • The needs of the district.  If the district needs a forecast that provides an approximate number of students in the system by grade level, then a system-wide cohort survival model does this effectively.  If the district needs a model that predicts the number of students at individual schools or the number of students by domicile, then other models are used.
    • The feeder-system design. If the feeder system (elementary to middle, middle to high) is clean, then a school-level cohort-feeder model provides an effective model. Even when the feeder system is not clean, if data can be found to indicate the percentage of students that move from one school to the next, this model can be used.  This model is not effective if there are large numbers of students who attend schools outside their domicile-base school or when there are special programs at a school that attract students from all areas of the district.

What is a Cohort Survival Ratio (CSR)?

  • This ratio is found by dividing the number of students in grade G+1 in year Y+1 by the number of students in Grade G in yearY.  A CSR that is greater than one means that in-migration is occurring in the system.  A CSR that is less than one indicates that the loss due to attrition and out-migration is greater than the in-migration that year in that grade. Many of the forecasting models used by OREd utilize this ratio.

What is a system-wide cohort survival model?

  • With at least four years of history, a system-wide cohort survival model uses an optimal weighted average of the cohort survival ratios (CSR) to move students from one grade to the next – year by year.

What is a planning segment?

  • Schools districts and attendance zones are far too large to analyze and use to create forecasts. OREd analyses the district carefully and divides it into planning segments which form the basis for analysis and predictions. Planning segments are created so that:
    • any one is homogeneous with regard to the type of residential development,
    • the boundaries of planning segments align with current attendance boundaries,
    • geographical features such as interstates, rivers, and lakes form the boundary.

What is a planning segment based model?

  • Districts are divided into planning segments. OREd uses parcel data and student data to analyze each planning segment. This analysis provides information about the type of residential development, the number of structures, the number of available lots, and the number of students generated for each residence (SGR), for example. Additional information is collected in a variety of ways that include personal interviews and a land-use survey. Using this information, OREd can predict future development and future new students. The number of newly developed structures each year, with the appropriate SGR, predicts the number of new students entering that planning segment because of new development. This information, along with returning students (calculated using a CSR) can predict the number of students in each planning segment.

What is a school-level cohort feeder model?

  • Using ADM data from previous years, cohort survival ratios by grade level at each school can be calculated and used to create school-level forecasts.


What is an SGR?

  • The student-generation-rate (SGR) is number that identifies the number of K-12 students generated by a particular type of residential structure. That is, if 100 new single family homes were planned for an area in which the historical SGR were 0.75, then one would expect 75 new K-12 students to come from those 100 new homes. The SGR depends on several factors, the most critical being the type of residential structure. A one-bedroom apartment complex might have an SGR of 0.10, a townhouse development; 0.40. OREd calculates the SGR for subdivisions by doing a parcel analysis on a subdivision and counting the geocoded students in that subdivision.

What information does OREd need from the District in order to construct a reasonable model?

  • In most cases, OREd will need ADM numbers for several years, building capacities, and a K-12 student data-base. The latter should include information to allow for a 98% to 99% match when the list is geocoded. The student data-base would also include any additional information (academic or demographic) that might be useful to the system when making decisions. In addition to information from the District, OREd usually obtains shape files from the planning department that oversees the District.

What is a Scenario?

  • OREd uses the term Scenario to mean a solution to a school district re-alignment problem. The scenario can be district-wide or within a subset of the District. Scenarios are usually done by level: elementary, middle, or high. An optimal scenario is produced by our GAMS software which assigns planning segments to schools based on given constraints in order to minimize the total distance traveled. The scenario can include existing school or the GAMS software can be programmed to select an optimal location for a new school(s). It is important to note that a computer-generated solution (Scenario) is a guide to Districts as they make decisions about student assignment or school locations. The District should also be guided by policies that direct decisions.

What District-Wide policies might influence final decisions about an optimal scenario?

  • Districts need to decide on priorities before making decisions on student-reassignment. Some of the policies that have to be prioritized include:
    • Solutions should address school-level utilization: achieving equitable utilization percentages for schools at the same level
    • Solutions should factor in distance: how far students must travel to get to school
    • Solutions might be influenced by feeder-patterns: in some cases, contiguous attendance zones are important
    • Solutions should impact as few students as possible: the number of students impacted by reassignment might be a factor in decision-making
    • Solutions might be influenced by student-balance metrics
    • Solutions might be influenced by the location of communities and/or municipalities: the intrinsic value of a community-based school may be more important than numbers

What is GAMS?

  • General Algebraic Modeling System (GAMS) is a high-level modeling system for mathematical programming and optimization. It consists of a language compiler and a stable of integrated high-performance solvers. GAMS is tailored for complex, large scale modeling applications typical in operations research, and allows you to build large maintainable models that can be adapted quickly to new situations.  OREd uses GAMS to find optimal solutions for school attendance boundaries and/or optimal sites for new school locations.

What is a Land Use Study?

  • OREd is able to provide Districts with a comprehensive Land Use Study (LUS) when necessary. A LUS involves stake-holder input via community interviews. Information obtained directly from district-level-decision-makers is an important ingredient to an effective model. The interviews are also beneficial in that they bring the stake-holders together and provide a means of communication that might not otherwise be present. Interviews help determine key trend indicators, factors that encourage or constrain growth, infrastructure issues (transportation, water/sewer,..), anticipated non-residential development, and anticipated residential development. Validated by a GIS-based analysis, the OREd can articulate growth potential in terms of attendance zones and even neighborhoods.

What is a What-If Attendance Zone Scenario?

  • OREd can produce a customized ScenarioBuilder for a District. The ScenarioBuilder allows Districts to modify optimal assignment scenarios on a planning segment level to assess what happens as a planning segment is reassigned.

K12 Forecast Comparison Line Graph