Research & Insights
Research is the feedback mechanism of the practice: outcomes correct assumptions, better assumptions improve models, and better models improve future decisions.
Feedback loop
Compare the forecast, assumptions and risks with the realized project outcome.
Only with appropriate rights and permissions, normalize lessons across engagements without identifying the client or project.
Use public records and other legitimate sources to study regional construction outcomes and recurring risk patterns.
Offer comparative products only when the evidence base is deep, normalized and governed enough to support the claim.
Client work does not automatically become research data. Any use beyond the engagement follows the applicable consent, confidentiality, anonymization and data-rights controls.
Why research belongs inside TRAVO
Replace generic assumptions with better-informed distributions where evidence supports doing so.
Learn which project conditions actually precede material cost, schedule or completion deterioration.
Convert what the practice learns into improved seminars, workshops, tools and decision processes.
Lines of inquiry
Research directions are framed as questions to test, not benchmark claims assumed in advance. Comparative products follow only when the supporting evidence is deep, normalized and governed enough to justify them.
01
Study cost-overrun, schedule-slippage, contingency and claim-emergence patterns using legitimate public sources and appropriately governed project data.
02
Compare the contingency held with the realized project outcome to understand where reserves proved insufficient, adequate or excessive.
03
Test whether recurring project conditions reliably precede material cost, schedule or completion deterioration, with analytical tools subordinate to expert judgment.
04
Study how plans and risk models should update as uncertainty resolves and new field information becomes available.
Evidence before positioning
Case studies are published only where the underlying engagement is supportable, publication rights permit it, and the result can be stated without implying outcomes that the evidence does not establish.
Decision
State the project stage, client perspective and decision trigger.
Method
State the information, model, sensitivity/scenario and probabilistic methods actually used.
Result
Report ranges, drivers, alternatives or other decision outputs without invented precision.
Outcome & learning
Where known and publishable, compare the forecast and decision with the realized outcome and lessons.
Executive brief · 2026–2031
By 2031, continuous, auditable risk intelligence becomes a condition of competing for major engineering programs — priced first through insurance and procurement, not regulation. Firms that stay on static risk documents face a growing “analog penalty.”
What leaders should take away
The bounded first step is a Project Risk Screen.
Request a decision-scoping call — principal@travo-advisory.com · New Jersey / New York metropolitan region.
Independent · principal-led · decision-focused
Selected peer-reviewed foundations
Selected publications are shown because they directly evidence the principal's published work in decision analysis, uncertainty, exploration planning, information value and formal updating — methodological foundations now applied through TRAVO. They are academic foundations of the method, not claims of client outcomes.
ASCE · 2007 · Part I
Karam, K.S.; Karam, J.S.; Einstein, H.H. Journal of Construction Engineering and Management, 133(5), 344–353.
Decision analysis for tunnel exploration, including assessing the effect of additional exploration before committing to it.
ASCE · 2007 · Part II
Karam, K.S.; Karam, J.S.; Einstein, H.H. Journal of Construction Engineering and Management, 133(5), 354–363.
Formal incorporation of uncertainty into exploration-planning decisions.
Georisk · 2014
Sousa, R.L.; Karam, K.S.; Einstein, H.H. Georisk: Assessment and Management of Risk for Engineered Systems and Geohazards, 8(3), 155–170.
Information collection, virtual exploration, Bayesian updating and decision analysis for determining when additional information is worthwhile.
