Fail Early

Order knock-outs by least effort and highest termination probability first.

category: Behaviour class: Routing Rules frameworkAspect: Behavioural view perspective: Control flow

Description

The knock-out pattern is a strategic and practical approach to improving operational efficiency. At the core of a typical business process are numerous conditions that must be met to achieve a desired outcome. However, if any of these conditions fail, it could lead to a premature termination of the process segment, termed a ‘knock-out’.

This pattern works by analyzing and sequencing the order in which these conditions are checked. The flexibility in this ordering allows for a smart strategy where the conditions with the most favorable balance of high termination probability and low effort are checked first, followed by the second most favorable, and so on.

In other words, we aim to verify the conditions most likely to fail and least costly to check before the others. This methodical ordering minimizes the overall cost and effort of executing the business process by terminating any potentially unsuccessful paths at the earliest stage.

It’s important to note that implementing the knock-out pattern requires a level of autonomy in the ability to order these checks. Also, while this approach generally leads to cost-efficient processes, it might potentially increase the total time taken to complete the process compared to simultaneous verification of all conditions. This is a factor to consider based on the specific needs and constraints of your business process.

The pattern is a special case of Resequence activities, and its natural counterpart in time-critical processes is Parallelize activities, where checks run concurrently instead.

Original heuristic (source excerpt)

A typical part of a workflow is the checking of various conditions that must be satisfied to deliver a positive end result. Any condition that is not met may lead to a termination of that part of the workflow, the knock-out. If there is freedom in choosing the order in which the various conditions are checked, the condition that has the most favorable ratio of expected knock-out probability versus the expected effort to check the condition should be pursued. Next, the second best condition, etc. This way of ordering checks yields on average the least costly workflow execution. There is no obvious drawback on this heuristic, although it may not always be possible to freely order these kinds of checks. Implementing the heuristic also may result in a (part of a) workflow that takes a longer throughput time than a full parallel checking of all conditions.

Reijers and Goverde (1999a) and Van der Aalst (2000b) mention this heuristic.

Van der Aalst (2000b) also gives quantitative support for its optimality.

Key Takeaways

  1. Try to fail cheap and early — Sequential checking of conditions, prioritizing those with higher knockout probability and lower checking effort, often leads to cost-efficient business processes.
  2. Flexibility required — Application might be limited in scenarios where the conditions’ verification order cannot be freely modified.
  3. Cost vs. Time — While this approach optimizes cost, it might extend the total process duration compared to a parallel verification of all conditions.
  4. Rank by ratio — Order checks by knock-out probability relative to checking effort, not by habit or form layout.

Performance considerations

QualityTimeFlexibilityCost080

■ This pattern   ■ Baseline

Adopting this sequential checking strategy usually leads to the most cost-effective operation of a business process on average. While there isn’t any clear downside to this approach, its application might be limited in scenarios where the order of checks can’t be freely adjusted. Furthermore, while this method optimizes cost, it may prolong the total processing time in some parts of the workflow, compared to a scenario where all conditions are verified simultaneously.

In Devil’s Quadrangle terms, cost is the clear winner: little effort is spent on cases that are eventually rejected, because they leave the process after the cheapest discriminating checks. Time is rated as slightly worsening, since sequential checks take longer than parallel ones for cases that pass every check. Quality and flexibility are neutral. If both cost and time matter, running the most selective low-effort checks first and the remaining ones in parallel is a common compromise.

Example

For example, a bank processing consumer loan applications could perform its checks in the order of the application form: income verification, employment check, collateral valuation and finally the credit bureau query. If the event log shows that the bureau query rejects a large share of applications, is automated and costs little, while the collateral valuation is expensive and rarely decisive, the bank could move the bureau query to the front and the valuation to the end. Rejected applicants would also receive their answer sooner.

Spotting the opportunity in event data

Framed along the FLAC rule types — fitness, location, attribute, constraint.

Fitness: processes in which a substantial share of cases ends in rejection or cancellation after several checks. Location: sequences of check, verification or approval activities that can end the case, followed by an end activity such as Reject. Attributes: knock-out rate per check (share of cases terminated directly after it), effort per check (processing time or cost), position in the sequence, effort already spent on rejected cases. Ranking checks by knock-out rate divided by effort and comparing that ranking with the observed order shows the opportunity. Constraints: data or legal dependencies that fix the order; checks that must be documented for every case.

Sources & references

  1. Lashkevich, K., Mediavilla Ponce, L. M., Camargo, M., Milani, F., & Dumas, M. (2023). Discovery of Improvement Opportunities in Knock-Out Checks of Business Processes. RCIS 2023 Summary on this site Publisher page DOI 10.1007/978-3-031-33080-3_23
  2. Reijers, H., & Liman Mansar, S. (2005). Best practices in business process redesign: an overview and qualitative evaluation of successful redesign heuristics. Omega, 33(4) Summary on this site Publisher page DOI 10.1016/j.omega.2004.04.012

Share this pattern

Enjoy these

Foundational free Patterns

Apply Integral technology

Elevate physical constraints by applying new technology

Apply triage

Consider the division of a general activity into two or more alternative activities

Appoint case managers

Assign a responsible individual for handling each case type

Assign cases

Let workers perform as many steps as possible for single cases

Automate activities

Consider automating activities

Automate for environmental impact

Implement automation in a sustainable way

Buffer information

Buffer external information and subscribe to updates

Combine activities

Combine small activities into composite activities

Consolidate Work

Collect similar work items and work in batches

Constraint-based task assignment

Allocate tasks considering business process execution constraints

← Back

Connected to

Isolate exceptions

Design business processes for typical cases and isolate exceptional cases from the normal flow

Resequence activities

Move activities to more appropriate places

What's Happening?

Dec 2023
New Pattern

Consolidate Work

Nov 2023
Oct 2023
Oct 2023
New Cheat sheet

Redesign Essentials

Sep 2023
Sep 2023
Aug 2023
New Pattern

Task delegation

Aug 2023
Aug 2023
New Pattern

Green Feature

Aug 2023