SBIR-STTR Award

Accurate projection of costs remains an intractable problem in the defense industry despite more than a decade of focus on affordability. The impact to DoD acquisition programs, from cost escalation and the cost of countering capability shortfalls, runs to billions of dollars annually. The objective of this proposal is to develop and demonstrate an ontology-based framework for requirements-based cost modeling. On-tologies offer a unique capability to manage the large knowledge bases and rule sets essential to understanding cost relationships in complex systems. InRAD's innovative approach uses a systems concept where domain-specific, plug-in compatible ontologies are populated with relevant knowledge for the products, processes, materials, and functions that must be considered in cost model. Product require-ments, captured using SLATE, DOORS, or similar tools, will be used to create a requirements-based life-cycle cost modeling framework structured to align with the ontology set. The resulting capability will enable engineering and production staff to draw on a far larger base of knowledge in projecting costs; re-fine estimates faster and with greater precision; better account for uncertainties and risks; and recalculate "on the fly" when requirements change or are traded off to define the optimal balance of cost and per-formance.

Keywords:
ONTOLOGIES, COST MODELING, PRODUCT REQUIREMENTS, PROCESS REQUIREMENTS, PRODUCT MODELS, PROCESS MODELS, ENTERPRISE MODELS

This SBIR Phase II, Requirements-Based Cost Models, will develop software to enable cost managers to extrapolate costs of complex systems with higher fidelity than current approaches. Identifying the true costs of transitioning technologies from concept to production is complicated by many factors, from the lack of a known cost basis from which to estimate new requirements to difficulties in identifying information that can support improved cost extrapolations. Based on an ontology-driven approach, the Requirements-Based Cost Modeling System (RBCMS) will be a web-accessible toolset that enables users to evaluate and document cost estimates, particularly in pre-proposal and proposal project stages. InRAD’s software will merge cost data with technical information to provide a foundation for determining costs for elements in order to (1) provide a basis for the derivation of cost drivers and their impact to total product cost and product requirements and (2) provide a systematic approach for cost managers to access highly targeted data to substantiate costs for which no historical or documented costs are known. The RBCMS will manage cost sessions to provide continual histories of changes and traceability to factors that impacted projections. As middleware, this system will supplement, not replace, current parametric and other cost modeling tools.

Keywords:
Requirements, Cost Models, Ontologies, Information Management, Knowledge Management, Information Retrieval, Text Mining, Cost Estimates