Jump to text
On this page
Copy and source
Published source confirmed at last check

Source changed 2026-11-16 21:18:08 UTC · snapshot created 2026-11-16 21:19:40 UTC · last check 2026-11-17 19:15:10 UTC

Practical entries

Start with the question that matters to the inquiry and the evidence you have. A design specifies what the observations must distinguish. The way you collect or recover material determines which comparisons you can make. Analysis, criticism and synthesis can support an answer, identify a limit or expose a premise of an earlier step that needs revision.

If the available results leave part of the research question open, use FPF B.5, Choose the Next Reasoning Contribution, to identify what another contribution would add. RMP.Reference:6 explains how to connect the research methods and when a change warrants revisiting earlier work.

Start with the contribution whose inputs you have. The Table of Contents covers other questions. The Architectural Reference explains why the methods belong here and what their combined use requires. When you seek help, ask for an explanation or a comment on your case in the language of your work.

Use RMP.1 — Decide Whether a Question Needs Research and When to Stop to check whether an applicable source already answers the question. If it does, give the answer, its source and its relevant limits. If a knowledge question remains, explain what the inquiry seeks to establish, how another researcher can examine the connection from question to result, and the scope of the contribution. A suitable existing protocol can support an inquiry into a disputed claim. An established rule can also settle an ordinary case in its own practice.

For a manager considering a second approver, see Should a second approver reduce customer waiting?. Event records can support a delay forecast under stated conditions. To estimate the effect of adding an approver, establish how the queues operate. The application shows the comparison across branches with independent queues, how a shared queue changes its interpretation, and what a qualified analyst can supply.

The examples use constructed cases with stated professional premises. Research can change the evidence available for maintenance, environmental protection or another decision. The authority to make the decision belongs to its designated holder.

RMP-SEAL — Separate routine qualification from a mechanism-validity inquiry

  • Situation: A seal engineer needs to know whether an accelerated test supports a proposed use in pump maintenance, but the relation between laboratory degradation and service conditions may be unresolved.
  • Question: What evidence can answer the engineer’s question? What happens to the conclusion if the comparison fails to distinguish the mechanisms or the laboratory result cannot be transferred to service?
  • First useful result or blocker: An applicable source answer or a finding under the existing qualification procedure. If a knowledge question remains, state it and assess whether a feasible inquiry can answer it.
  • Start with: RMP.1 for an unsettled question. Use RMP.2 — Choose a Criticism-Bearing Research Design or Stop directly when a compatible question, source account and useful stopping condition are already supplied.
  • Stop or return: Give the answer the evidence supports. Use RMP.1 if the question has changed and RMP.2 if a design premise has failed. A decision about maintenance belongs to the authority responsible for it.
  1. Establish what is known about the mechanism. In the Pump-P4 application, a qualified acceleration relation, test method and acceptance criteria may suffice for the engineering or assurance task. Evidence that the proposed acceleration changes the degradation mechanism may also answer the question directly. If plausible alternatives remain, use RMP.1 to specify what would support, narrow or defeat the proposed relation for the stated material and operating conditions.
  2. Construct a comparison that can distinguish the mechanisms. In the mechanism construction, two reference specimens have the same overall wear index and different degradation mechanisms. The index cannot tell the mechanisms apart. A qualified procedure for measuring a distinguishing signature lets the researcher compare accelerated and field-representative specimens under the required material and damage conditions. With RMP.2, the researcher can use sufficient evidence, select a feasible design, identify a missing result that blocks a needed selection, or give a reason to finish without a new study. Unknown marker discrimination blocks selection if the choice depends on it. Two necessary tasks of four to six non-overlapping investigator-days each could take eight to twelve days. A ten-day allocation covers some combinations, but does not ensure completion.
  3. Carry the intended comparison into specimen handling and measurement. Once the marker, measurement procedure, material plan, safety conditions and access are suitable, use RMP.3 - Operationalize the Research Design without Replacing the Question to specify specimen identity, allocation, observation timing and how the readings will enter the comparison. Three readings of one seal give information about that one specimen. If a specialist establishes that the marker cannot distinguish the mechanisms, return to RMP.2 with that result. Using a wear score instead would require evidence that it answers the mechanism question.
  4. Interpret the observations that survived. With RMP.4 - Preserve a Criticism-Ready Research Trace, trace the handling damage to specimen P4-07, its missing reading and the decision to exclude it. Use RMP.5 - Analyze and Criticize the Research Claim under Its Method to examine what follows for the inference. The missing reading leaves the direction of bias open. Reproducing a simulated curve confirms its reproduction, while a claim about field performance needs evidence relating the model to field conditions. Use RMP.6 - Choose a Credibility Test That Bears on the Research Claim, or Answer without New Testing to assess whether an attainable mechanism or field comparison would be worth its cost. A selected comparison still awaits results.
  5. Combine estimates of the same quantity. In the corpus branch of the case, use RMP.7 - Synthesize a Research Evidence Corpus to trace A and A2 to one test, B to an independent compatible test, and C to service replacement under different conditions. A2 adds detail about the first test. The two independent estimates of the laboratory mean-lifetime difference are 14 hours with standard error 4 and 8 hours with standard error 3. Under the justified common-effect model, inverse-variance weighting gives 10.16 hours with standard error 2.4. The normal approximation gives a 95% interval of about 5.46–14.86 hours. C can answer a separate service question.
  6. Revise the claims that rely on the result. Use RMP.8 - Use Research Results to Retain or Revise an Episteme’s Claims and Return Bounded Evidence to find the dependent claim. In the model example, a separate analysis supports a fitted wear relation for the dry laboratory. Its transfer to wet service remains unproved after the synthesis. Retain the dry-laboratory fit, withdraw the claim of demonstrated transfer and the service replacement-time prediction that depends on it, and tell the reliability practitioner which premise is missing. The engineering wear limit and any maintenance decision keep their own grounds.

A supplied result can support analysis, a credibility comparison or synthesis directly. If the common-effect premise fails, reconsider what the synthesis establishes. Qualified new field evidence can reopen the transfer question without repeating the laboratory calculation.

RMP-CITIZEN — Investigate a stream anomaly without an affiliation gate

  • Situation: Residents have downstream salinity records, a proposed explanation and material from other catchments. A comparison they intended to make becomes unavailable.
  • Question: What do the surviving observations establish? How should the briefing reflect the loss of the intended comparison?
  • First useful result or blocker: The observed pattern and the comparison or source relation still missing. If the question qualifies for research, identify a suitable inquiry or explain what can be concluded without a new study.
  • Start with: RMP.1 if monitoring and research remain confused. With a selected field design, start at RMP.3; with inherited material, start at RMP.4.
  • Stop or return: Report the supported observations and explain the limit on attribution. Use a qualified measurement result from the laboratory or another competent provider. Site access, use of community data and decisions about the environmental response depend on their respective permissions and responsibilities.
  1. Check what the monitoring rule answers. In the ResidentStream-S17 application, sampling under an established threshold and response rule can finish with the requested monitoring result. If the question concerns the cause, ask what would distinguish road-salt transport, seasonality and sensor drift. The question and the work establish whether this is research, whoever performs it.
  2. Relate the measurements to the design. Given qualified measurement, safe access and suitable comparison periods, use RMP.3 to specify the site, period, hydrological conditions, sample and procedure. A laboratory can supply the measured value. The residents can then connect it to their site-period comparison.
  3. Account for the missing thaw observation. Suppose the reference reach becomes inaccessible during the informative thaw and observations downstream continue. Use RMP.4 to record the downstream observations, the missing site-period and why it was missed. A later dry-weather reference reading concerns different conditions. If the design needs the paired thaw observation, return that loss to RMP.2. Use the surviving record for any narrower claim it supports.
  4. Report the pattern and the limit on its explanation. With RMP.5, the residents may be able to describe the downstream pattern while its source remains open. They can finish with that finding. If a later matched observation could answer a remaining question, use RMP.6 to assess whether it is feasible and worth obtaining.
  5. Use other catchments for the questions they cover. With RMP.7, compare the other studies under the conditions in which their observations were made. A similar chart elsewhere leaves the missed local event unobserved. Use RMP.8 to correct any unsupported local attribution in the briefing, keep the downstream finding and tell the recipient what is still unknown about the cause.

In the historical application, several histories repeat one memoir. An authenticated record dates the decision to close the factory before the protest said to have caused it. This chronology rules out that explanation for the decision. Its actual cause and the protest’s possible effect on later implementation remain open. With RMP.8 the researcher can give the museum editor the correction and its exact scope.

RMP-CORRECTION — Repair an analysis and return its consequence without discarding a suitable method

  • Situation: A completed analysis contains a calculation defect, and a later change in the observations may also challenge the method’s fit.
  • Question: Is the defect in the computation or the method? Under what conditions would a comparison of methods answer a further question?
  • First useful result or blocker: A verified correction and any resulting change in the analysis or claim its users rely on. Under different conditions, the researcher may be able to select a method or explain which question about its suitability remains open.
  • Start with: RMP.9 — Decide Whether and How to Improve Research Methods and Practice, using the actual difficulty and receiving conditions.
  • Stop or return: Keep the suitable method after repairing the computation. If conditions have changed, report a proposed comparison as a plan until results are available.
  1. Locate the calculation error. In the calculation application, the method for the independent observations 2, 4, 6 and 8 specifies sample variance with denominator n−1. The mean is 5 and the sum of squared deviations is 20. Dividing by n gives the reported 5. The specified denominator gives 20/3. With RMP.4 the researcher can trace the computation and find the analyses that used it.
  2. Correct and verify the affected result. Use RMP.9 to repair the denominator and retain the method. With RMP.5, recalculate the affected analysis. With RMP.8, return any changed uncertainty or conclusion to those who relied on it. The known-answer check settles this implementation defect. Earlier analyses that did not use the faulty calculation retain their support.
  3. Reassess the method for a different target. Now the programme wants an equally weighted mean across independent sites. The numbers of dependent observations per site are unequal. Pooling all observations gives the sites different weights. Use RMP.9 to compare methods for the site-level quantity. A qualified statistician, existing simulation code and two investigator-days make it feasible to compare a t interval from independent site means with an interval obtained by resampling whole sites. The team has selected this comparison. Which procedure is suitable remains open until the comparison or other qualified evidence answers it. Restrict the unsupported uncertainty claim and reconsider the comparison if its cost, value, access or staffing changes.