Research · Selected work

Schools as Critical Infrastructure

A school-infrastructure decision-support project prioritizing community services through the RURU framework, weighted scoring, stakeholder mapping and an interactive decision-space prototype.

The framework covers 20 services in seven categories, with separate equity and governance outputs.
Diagram in text
Define services
Meals, connectivity, childcare and other needs.
Structure evidence
Translate surveys and interviews into inputs.
Build RURU profiles
Read resources, urgency, rules and uncertainty.
Add decision context
Examine equity, stakeholders and authority.
Compare options
Review the worked example and rulebook.

Interactive tools

  • RURU decision map

    Adjust the decision dimensions and explore entries in the schools-as-infrastructure framework.

  • 3D RURU prototype

    Explore the school-service map in three dimensions using the research prototype.

Approach and contribution

See the results

30-second case brief

A transparent decision framework for keeping essential school services accessible during a crisis.

Problem
Compare urgent community needs with district capacity, changing rules, uncertainty and unequal access.
My contribution
Decision-matrix development, systems architecture, scoring structure, stakeholder models and interactive prototype within the Drexel–Temple collaboration.
Outcome
A scoring rulebook and worked demonstration spanning 20 services across seven categories, with separate RURU, equity and governance outputs.

ApproachResearch evidence → standardized inputs → service profiles → equity and stakeholder interpretation → traceable decisions.

Project description and responsibility

Context
Spencer Foundation research · Drexel University and Temple University
Contribution
Decision-support workstream · systems analysis, matrix development and prototype structure
Research basis
Research-team superintendent survey and qualitative interviews in Pennsylvania
Core services
Food and meal distribution · information and communication technology · childcare
Decision unit
District × service × crisis time window
Outputs
RURU profiles · equity scores · stakeholder/governance indicators · worked service comparisons

Interactive tool · RURU 4D map

Map services across four decision dimensions.

Select an entry, adjust its position, or add a new one. Horizontal position shows urgency; vertical position shows available resource capacity. Larger, lighter points show greater uncertainty; thicker borders show more regulation.

Editable browser model based on the project’s RURU framework. It is a decision-structuring aid, not a risk prediction.

Map convention: Resources means available capacity (0–10), following the original explorer. The newer rulebook uses a different Resources score for implementation burden (0–1). These editable map coordinates are demonstration inputs, not calculated rulebook scores.

More available capacityLess available capacityLower urgencyHigher urgency
Smaller / darker: lower uncertaintyLarger / lighter: higher uncertaintyThicker border: more regulation

The original exploratory map uses Resources as available capacity on its vertical axis. The newer scoring rulebook below expresses Resources as implementation burden. The map’s editable values are demonstration inputs.

Open the original 3D RURU prototype
3D RURU prototype

Schools are more than places of instruction

Keep essential services working when ordinary routines stop

During the COVID-19 disruption, school districts had to organize meals, devices, internet access and childcare while their buildings and teaching routines changed. The Spencer project examined schools as critical social infrastructure: facilities, staff, transport, institutional knowledge and community relationships working together to support families.

That perspective changes the decision problem. Keeping a service available is only one part of success; families also need a practical way to reach and use it. Food delivery depends on procurement and distribution, digital learning depends on both devices and connectivity, and childcare depends on staffing, space and operating rules. The decision-making system brings these different conditions into an explicit comparison.

My contribution was the decision-support workstream within the Drexel–Temple collaboration. I developed the matrix structure, systems architecture, scoring approach and stakeholder views to make the reasoning behind a service decision visible.

Three core services and their implementation questions
ServiceConditions the decision must represent
Food and meal distributionMeal provision, staffing, procurement, transport and family access
Information and communication technologyDevices, broadband, technical support and usable access at home
ChildcareDemand, eligible users, staffing, facilities and operating requirements

Spencer project description · research objectives; DMS executive summary · purpose

Different pressures need different responses

Translate the research team’s findings into decision variables

The team’s survey and interview synthesis found that districts expanded food, technology and health support during the initial disruption. Leaders mobilized existing staff, institutional knowledge and community relationships, while implementation was shaped by funding, staffing, distribution logistics, connectivity and changing guidance. Intermediate Units helped connect districts with peer networks, health expertise, legal interpretation and state agencies.

The synthesis also identified unmet needs in broadband, home health and wellness visits, and childcare for teachers’ children during school hours. These findings explain why a single measure of service availability is insufficient. Capacity, need, access and coordination describe different parts of the problem, so the DMS represents them separately.

The research informed the choice of variables and the interpretation of each decision layer. The worked scoring demonstration then made those choices explicit through documented directions, weights and output definitions.

Research findings and their implications for the framework
Finding in the team’s synthesisDecision-system response
Expanded food, technology and health supportsCompare service needs and implementation conditions
Broadband and childcare remained important unmet needsRepresent unmet demand and service-specific access gaps
Staffing, logistics and infrastructure affected deliverySeparate operating constraints from capital and technology constraints
Guidance and demand changed during the crisisRetain regulatory context and uncertainty as distinct dimensions
IUs supported inter-district and agency coordinationModel stakeholders, authority and operational dependencies

Research-team Summary of Findings for Interviews and Survey · survey and interview synthesis

Make the route from evidence to output auditable

Separate stable definitions from changing crisis conditions

The full architecture separates a static reference layer, a dynamic evidence layer and a computational layer from the user interface and an auxiliary explanation layer. Service definitions and scoring rules belong in the static layer. Time-varying staffing, demand, regulations and access conditions belong in the dynamic layer. The computational layer turns those inputs into the decision profiles.

This separation gives each number a context: the district, service and crisis time window being assessed, the source information, and the scoring rule used. The completed rulebook establishes the analytical core of that architecture. The diagram shows how a future operational platform could organize data updates, interpretation and scenario exploration around it.

The proposed full architecture separates data, calculations and presentation. The scoring rulebook and worked calculations form its documented analytical core; the interface and explanation layers describe the intended operational design.

DMS executive summary · June 26 revision

Standardize scoring direction and preserve the meaning of missing inputs

The rulebook maps ordinal inputs onto a common zero-to-one scale and standardizes their direction. Higher values generally indicate greater need, burden, constraint, uncertainty, vulnerability or inequity. This allows variables measured on different source scales to contribute to interpretable service profiles.

The treatment of applicability is equally important. A variable that does not apply to a service is excluded, and the remaining weights are renormalized. Missing evidence is flagged as a data-quality or confidence concern rather than silently becoming a zero. Service-specific weighting then reflects which operating resources and capital assets matter most to that service.

The resulting calculation can be followed from raw evidence to standardized variables, thematic clusters and final outputs. That traceability is the basis for comparing alternatives and revisiting an assumption when conditions change.

Rules that preserve the interpretation of an output
Input conditionTreatment in the current rulebook
Different source scalesNormalize to a common 0–1 range
Different scoring directionsAlign higher values with the defined burden or need
Variable does not applyExclude it and renormalize the remaining weights
Relevant evidence is missingFlag missingness and confidence rather than substitute zero
Services depend on different resourcesUse documented service-specific operating/capital weights

DMS executive summary · June 26 revision

Keep urgency and implementation conditions visible

Read the profile before choosing a priority

RURU describes a service through Resources, Urgency, Regulations and Uncertainty. In the current rulebook, Resources means implementation burden: staffing, finance and supplies are assessed alongside physical infrastructure and technology. These groups can receive different weights for food distribution, broadband access or another service.

Urgency describes how immediate and consequential the need is. Regulations captures the rules and guidance affecting action, while Uncertainty captures incomplete information and unstable conditions. A high score in each dimension therefore calls for a different response: capacity support, timely action, compliance or clarification, or adaptive planning and additional evidence.

The original spatial matrix shows Resources, Urgency and Uncertainty on three axes, with Regulations represented by color. Tables and paired two-dimensional views offer simpler ways to read the same profile. An optional integrated score can support a comparison while the underlying dimensions remain available for interpretation.

The original four-dimensional profile visualization. Point positions and colors are illustrative design examples, not measured district priorities.

DMS executive summary · June 26 revision

Follow meal distribution from inputs to an urgency score

The rulebook’s worked meal-distribution example assigns raw scores of 5 for service importance, 5 for unmet need, 4 for operational difficulty and 5 for community dependency. Applying (raw score − 1) / 4 gives 1.00, 1.00, 0.75 and 1.00. The demonstration weights are 0.30, 0.30, 0.20 and 0.20 respectively.

The resulting urgency score is 0.30 × 1.00 + 0.30 × 1.00 + 0.20 × 0.75 + 0.20 × 1.00 = 0.95. This is the source document’s worked demonstration, with its stated inputs and weights. It shows exactly how an output is obtained and remains separate from the editable 0–10 map coordinates above.

Meal distribution: the source rulebook’s worked urgency calculation
InputRaw score (1–5)Normalized score (0–1)WeightContribution
Service importance51.000.300.30
Unmet service need51.000.300.30
Operational difficulty40.750.200.15
Community dependency51.000.200.20
Urgency score1.000.95

DMS Prototype Scoring Rulebook and Worked Example V1.3 · Tables 6, 7, 10 and 12

Ask who can use the service and who makes it work

Separate community vulnerability from service-specific disparity

The Equity layer distinguishes structural vulnerability from the access gap attached to a particular service. A district’s economic or connectivity conditions may affect many services, while delays, unmet demand and unequal participation can be concentrated in a particular program or population.

The prototype sets out additive, multiplicative and hybrid ways to combine these components. It also distinguishes within-district comparisons of services or populations from across-district comparisons of standardized profiles. The original equity graphic makes the units of analysis explicit: one district’s service-level needs on the left and district-level profiles for regional coordination on the right.

Two applications of the Equity layer: compare services and populations within a district, or compare standardized district profiles for regional coordination. The graphic presents the proposed analytical structure.

DMS executive summary · June 26 revision

Connect a service decision to authority, interest and values

The stakeholder model extends a conventional power–interest matrix with service-specific values and operational roles. Power may come from formal authority, funding, regulation or an operational dependency. Interest may come from reliance on the service, direct impact, advocacy or involvement in delivery.

The food-service example places funding compliance, meal provision, routine distribution and delivery support in different engagement positions. This reveals why technical feasibility alone does not describe implementation. The same service can require funding approval, staff coordination and communication with families, each involving different actors and values.

These inputs also support governance indicators for coordination complexity, pressure, conflict and dependency. Keeping them alongside the RURU and equity outputs makes the decision profile useful for both service planning and stakeholder engagement.

Worked food-service stakeholder example. The quadrants connect power and interest to engagement strategies; the boxes identify relevant actors and values. Positions are prototype examples.

DMS executive summary · June 26 revision

A documented framework spanning 20 services

Make the decision logic inspectable across different service needs

The completed deliverable is a scoring rulebook and worked demonstration covering 20 services across seven categories. It establishes the service and input catalog, scoring directions, normalization, missingness and applicability rules, RURU calculations, equity alternatives, stakeholder/governance models and integrated output tables.

The demonstration carries the same structure beyond the original food, ICT and childcare focus. Its values are worked examples that make the scoring logic inspectable, including the effects of resource weights and alternative equity formulations. The outputs retain the operational, access and governance reasons behind a comparison, rather than requiring one score to stand in for all of them.

The interactive tools on this page provide a way to explore the matrix and change a service profile. Together with the documented calculations and source presentations, they make the project’s contribution concrete: a reusable, traceable structure for discussing service priorities, implementation constraints and equitable access with school decision makers.

Completed decision-support outputs
DeliverableWhat it makes visible
Service and input catalogA common structure for 20 demonstration services across seven categories
Scoring rulebookDirections, normalization, applicability and documented weights
RURU profilesNeed, implementation burden, regulation and uncertainty
Equity alternativesStructural vulnerability and service-specific disparities
Stakeholder/governance modelsAuthority, interests, values and coordination dependencies
Worked comparisons and scenario optionsHow inputs and assumptions lead to interpretable outputs

DMS executive summary · June 26 revision

Contact

Do you have a question or want to chat?

Analytics preferences

Optional Google Analytics uses cookies to measure page visits and engagement. You can browse the entire portfolio with analytics off.

Analytics is off in this browser.

Privacy details. Change your choice here at any time. Your preference is remembered for up to 180 days.