The Exceptional Practice · Daily Brief10 November 2026

Singapore Airlines Makes Time to Get Better. Your Clinic Can, Too.

Five everyday ways to split your team’s protected time between practice, one fix and one new idea, and to keep it in busy months and good ones.

What you’ll take away
The problem
The time a team sets aside to practice, fix problems and try new ideas is often the first thing a busy month takes, and a quiet month takes the rest.
What to remember
Decide how your team’s protected time is split before the month starts: practice, one fix, one new idea. Then keep it, in good months too.
What to do differently
Block next month’s team hour before booking opens and write its three parts in the note. Move it within the month if you must. Never cancel it.

The time a team sets aside to get better is often the first thing a busy month takes. It’s the 25th, and the practice manager is building next month’s schedule. The Mohs list is overflowing, and the one open slot is the monthly team hour. It goes to patients, with a promise to train “later.” So nobody walks through what to do when a patient faints in the procedure chair. Prior authorizations go missing on surgery day again, and the front desk works around it again. The scheduler’s idea from last month never gets an answer. The month after is quieter, so the hour seems unnecessary and goes too. Nobody chose to stop getting better. Nobody decided at all. One Singapore Airlines manager described what deciding looks like.

The crew member asks: “Still practicing, after all these years?”

The team lead says: “Always. Once more, from the top.”

In a bright training room beside a mock aircraft cabin, with two rows of empty teal seats on a low platform and an oval window whose shade is closed, an experienced cabin crew member in her forties, in a plain teal jacket and dark skirt, holds a small wooden tray with one white cup in both hands and smiles, a little amused. Facing her, the team lead, a woman in her fifties with a grey-streaked bob and a plain navy blazer, holds out an open hand, palm up, toward the crew member, calm and encouraging.

The biggest share goes to people. One SIA manager described 40% of resources going to training and invigorating staff.

The agent says: “This handover step trips us up every week.”

The team lead says: “Then it’s the one we fix this month.”

At a long wooden table in the same training room, a ground-services agent in a plain grey jacket sits and points to the third of five boxes joined by arrows on a large sheet of paper with no words on it, looking up at the team lead, a little worried. Beside her, an engineer in plain dark-blue overalls leans on the table with both hands, studying the sheet. At the far end of the table, the team lead holds a pen and smiles at them, decided. A corkboard of blank cards hangs at the far right.

A share for fixing processes. In the same description, 30% went to reviewing processes and procedures.

The staff member asks: “Did anyone look at my idea?”

The team lead says: “Every idea gets looked at. We’re trying yours.”

Beside a corkboard of twelve blank index cards in the same training room, a young staff member in a plain light-blue shirt holds one blank card in both hands and looks at the team lead, hopeful. The team lead rests one hand on the edge of the board and smiles back at him warmly.

A share for new ideas. The last 30% went to creating new product and service ideas.

Illustration of the practices described below, not of real people or events. Dialogue is illustrative.

One Singapore Airlines manager described a rule for where the effort goes. In a teaching case by business-school researchers Wirtz and Heracleous, first published in 2016 and revised in 2021, a senior manager in charge of staff development at Singapore Airlines (SIA) describes the 40-30-30 rule. It sat inside the airline’s Transforming Customer Service program and covered people, processes and products together. In her account, 40% of SIA’s resources went to training its people and keeping them energized, 30% to reviewing how the work gets done and 30% to coming up with new products and services.1 Every kind of getting-better work has a named share, and the biggest one goes to people.

The biggest share goes to people, and the training never finishes. In the same case, cabin crew recruits of that time took 15 weeks of intensive training.1 It quotes former chief executive Cheong Choong Kong saying that, for the company, “training is forever,” whatever a person’s age.1 In the year to March 2026, SIA and its low-cost airline Scoot reported that each employee averaged 76 hours of training, and cabin crew 85.3

Staff ideas had a standing channel, and every one was evaluated. In their study of the airline, Heracleous and Wirtz describe Staff Ideas in Action, a suggestion scheme that had run for almost 30 years with a focus on cutting costs and improving efficiency. In 2007 and 2008, staff contributed “nearly 5,948 ideas.” Every one was evaluated, and over 40% were put into use.2

The time can have a fixed place in the week. Scoot, SIA’s low-cost airline, introduced SPARK Time: two hours set aside each Thursday that its office staff can use to learn on their own, share what they know and talk about leadership.3 A standing block doesn’t have to win its place each week. It already has one.

Why decide ahead? Getting better pays off later. The day’s work pays off now. Management researchers Repenning and Sterman argue that time put into improvement brings the more lasting gain, but only after a delay. So under pressure, improvement is the time that gets cut, and a short push to work harder turns into the normal way of working.6 The pull doesn’t stop when things improve. At DuPont, after a maintenance program’s gains, one of the team that built the model warned that people “will be taken away from the effort and the problems will go back up.” Later, company-wide cost cuts ordered from above brought downsizing that weakened the program.6 A split decided before the month starts is one calm decision, not a fight with every busy day. Keeping it in the good months is the harder half.

Principles of Excellence

5 Exceptional Practices for your Clinic

  1. Set the split before the schedule opens. The SIA manager’s rule gave people, processes and new ideas each a share. A split settled once, in a calm moment, doesn’t have to be won again on every busy day. When the overflow Mohs list starts eyeing next month’s team hour, the practice manager takes it to the lead physician before booking opens. Together they block the hour and write its three parts in the appointment note: about 24 minutes to practice, 18 to fix one process and 18 for one new idea. If the lead physician judges that a patient needs that time urgently, the patient is seen at once, and the practice manager rebooks the hour within the same month.
  2. Give practice the biggest share, and rehearse the rare moment. In the manager’s rule, people got the largest share. A rare event gives nobody the chance to learn it on the job, so its rehearsal has to be scheduled. When the nurse realizes nobody is sure of their job if a patient faints in the procedure chair, the practice minutes go to a walk-through at the chair, led by the people your clinic’s written emergency protocol names. The team rehearses each role the protocol assigns: who stays with the patient, who positions them, who brings supplies and who decides when to call emergency services. Anything missing or unclear goes to the next fix minutes. The walk-through follows the written protocol and adds to required certification. It never replaces it.
  3. Spend the fix share on one lasting repair. The manager’s rule kept 30% for reviewing how the work gets done. Fixing a cause takes time once. Working around it takes time every week. When records staff find two pathology reports that sat unrouted for days because the covering-clinician field was blank, they report them at once to the person your clinic’s results protocol names. That same day, under that protocol, a clinician reviews both reports and makes sure both patients are told. The fix minutes are for the lasting repair: one change, such as making that field required, with an owner, a start date and the sign-off your protocol requires.
  4. Ask for ideas, and answer every one. In 2007 and 2008, SIA’s staff scheme evaluated every idea it received. An answer shows people the idea time is real. When a scheduler suggests texting Mohs patients what to bring and how long the day may last, the practice manager logs it, decides yes, not now or no, and opens next month’s idea minutes with the answer. Before the text is sent, any clinical instruction in it, such as on eating, medicines, wound care or preparation, is approved by the clinician your clinic’s protocol names.
  5. Move it if you must. Never cancel it, even in a good month. Scoot’s learning block sits in the same two hours every Thursday, so it never has to win its place. At DuPont, the pull to take people off improvement came after the problems went down. When the lead physician wants to book one patient into the team hour “just this once,” or a quiet month makes the hour feel unnecessary, the practice manager names the rule they agreed: the hour runs in good months too, and it is paid time for every role. A patient who needs urgent care is seen at once. Only the practice manager rebooks the hour, and only within the same month.

Explore the Evidence Further

Explore the Evidence Further. All 18 sources, with methods, findings, limits and clinic applications

Eighteen sources, grouped by kind of evidence: accounts of how Singapore Airlines described and organized its training, process work and staff ideas; what the airline reports now and reported in a hard year; research on why protected improvement time disappears; field studies of problems worked around and of protected learning time in health care; and studies of rehearsal and simulation. Each row gives the design and its main caution, the key finding, why it matters and a way to apply it. Study names link to the original sources, and the full references close this section. None of these sources tested a fixed split of team time, in an airline or in a clinic, so every clinic application is our proposal, not a tested intervention.

The Singapore Airlines practice: a teaching case and a ten-year study

Study or sourceSetting, design and sampleKey findingWhy it mattersHow to Apply It to Your Derm Clinic
Wirtz and Heracleous, “Singapore Airlines: Managing Human Resources for Cost-Effective Service Excellence”1Teaching case: A business-school teaching case (© 2016, revised 2021) that draws on 2008-2010 publications and 2015 interviews; mostly company voices quoted by researchers. Read in full. The 40-30-30 figure rests on one statement by an unnamed senior manager responsible for staff development. The statement is undated and may predate 2016, its unit, resources, is not defined, and no scope or period is given. No other primary statement of the rule was found. The case gives no audit and no outcome data.The manager describes a 40-30-30 split inside the Transforming Customer Service program, which involved staff in five operational areas: 40% of resources to training and energizing people, 30% to reviewing how the work gets done and 30% to developing new products and services. She adds that pilots and station staff also deal with customers at times. The case describes intensive 15-week courses for new cabin crew recruits, which its authors describe as the industry’s longest and most thorough without giving comparison data; this is the case’s historical figure, not a statement of today’s recruitment requirement. It quotes former chief executive Cheong Choong Kong on lifelong training and chief executive Goh Choon Phong pledging to keep investing heavily in training even in the hardest times; neither quote is dated, and the case gives no data on training spending in a downturn. The same case says functional training and retraining take nearly half of what SIA spends, a second undefined figure that does not match the 40-30-30 split. It describes cabin crew in wards of about 180 led by ward leaders, with check trainers who fly with teams, and an HR model of five linked elements, derived from senior-management interviews: stringent selection, extensive training, service delivery teams, frontline empowerment, and rewards and recognition. On the employee side, the authors report pay about average for Singapore and low by world standards, union disputes including a 2007 pilot pay dispute over the A380 that the Industrial Arbitration Court resolved, and study questions noting that five-year renewable crew contracts would draw disapproval in the US and Europe.It is the only primary record of the split. It places training inside a wider people system and states some of that system’s costs to staff.Proposed: name each kind of getting-better work and give it a share before the month starts, without treating SIA’s numbers as a target.
Heracleous and Wirtz, “Singapore Airlines: Achieving Sustainable Advantage Through Mastering Paradox”2Qualitative case study: A peer-reviewed single-firm study (2014) based on 31 in-depth interviews with senior and middle managers and cabin crew, and repeated site visits, from 2001 to 2011. Selected methods and innovation sections read. The authors have a long engagement with SIA, and there is no comparison group.Staff Ideas in Action had run for almost 30 years, focused on cutting costs and improving efficiency, with a CEO Top Suggestor award. In 2007-2008 staff contributed “nearly 5,948 ideas” at Group level; every idea was evaluated and over 40% were implemented (about 2,379, recomputed). A small central product innovation department, staffed by people rotated in for two or three years at a time, handled major projects such as the A380 cabin through a structured process, while individual departments ran their own continuous improvement, funded from their own budgets.It shows process improvement and staff ideas given a standing home, rather than left to spare time.Proposed: answer every staff idea with yes, not now or no, at a set time each month.

What the airline reports now, and in a hard year

Study or sourceSetting, design and sampleKey findingWhy it mattersHow to Apply It to Your Derm Clinic
Singapore Airlines, Sustainability Report FY2025/263Company report: SIA’s own report for the year from April 2025 to March 2026, published 25 June 2026, covering SIA and Scoot (21,667 employees at 31 March 2026). SIA states that it did not seek external assurance for the report. Averages are total training hours divided by employees on 31 March. Relevant pages read.SIA and Scoot reported spending $68 million on staff learning and development (the report prints the currency only as “$”) and 76 hours of training per employee on average, with cabin crew averaging 85. The five-year table shows training hours rising and falling: cabin crew averaged 59 hours in FY2021/22, then 81, 109, 80 and 85 in the four years after; all employees 80, 67 and 76 in the last three years; managers and above 36 falling to 25 in the last year. From FY2025/26 Scoot counts crew training hours a new way, so the latest year is not like-for-like with earlier ones. Scoot introduced SPARK Time, two hours set aside each Thursday for office staff to learn on their own, share knowledge and talk about leadership; the report gives no start date, uptake or outcome data, and it applies to office staff.Training is still heavy and reported today, and SPARK Time is a current example of protected time with a fixed place in the week.Proposed: give the team hour a standing slot, like an appointment, and move it rather than cancel it.
Singapore Airlines, Annual Report FY2020/214Company report: The company’s own Group-level account of the year to 31 March 2021, in Singapore dollars. Selected pages read. The report does not reconcile its job-loss narrative with its year-end staff table, and the research found no training-spending figure in it.The Group reported a net loss of S$4,271 million. Before any job cuts, a recruitment freeze from March 2020, early retirement, voluntary release for cabin crew and voluntary no-pay leave reduced headcount by around 1,900 positions. After a September 2020 announcement of 4,300 positions to be cut across the Group, and further pilot pay cuts, the final number of involuntary job losses was about 2,000. That figure is involuntary losses only, not the whole fall in headcount. Salary cuts started with management and were extended to all employees.It is the hardest recent test of the pledge to keep investing in people through difficult times, and it gives no training-spending figure that would settle it.Proposed: decide in a calm month what protected time a hard month keeps, before the hard month arrives.
Singapore Airlines, Cabin Crew careers page5Company web page: An undated recruitment page, accessed 10 November 2026. Only its contract terms are used here.New cabin crew are based in Singapore on a first contract of five years, which can be extended depending on performance. The contract model that the teaching case’s study questions expect would draw disapproval in the US and Europe is still current.Heavy training sits alongside employment terms that staff and observers may see critically; training is not evidence of better working conditions.Proposed: make protected time paid time for every role, including hourly staff.

Why protected improvement time disappears

Study or sourceSetting, design and sampleKey findingWhy it mattersHow to Apply It to Your Derm Clinic
Repenning and Sterman, “Nobody Ever Gets Credit for Fixing Problems that Never Happened”6Management research: California Management Review, 2001. A system-dynamics model with field examples from US manufacturing, product development and other firms, chosen by the authors; not a controlled study and not a test of any fixed three-part split. Mechanism and DuPont passages read.Time put into improvement yields more lasting gains than extra work, but only after a delay, so under pressure organizations cut it and fall into a self-reinforcing capability trap, in which a temporary push to work harder becomes the routine. In their study, successful improvement initiatives were the exception; even where early results came, cost and schedule pressure drained resources away. At DuPont, a 1991 in-house benchmark found maintenance spending 10-30% higher than industry leaders relative to plant value, and uptime some 10-15% lower; the improvement team’s model concluded that escaping would require a worse-before-better period of added planned maintenance, training and parts quality, a conclusion of the model rather than an observed result. After the program’s gains, someone on the modeling team warned that staff would be pulled off the work once problems fell, and company-wide cost cuts ordered from above then brought downsizing that weakened the program and limited its expansion.It explains why improvement time loses to the day’s work, and it shows protected effort eroded after it succeeded.Proposed: keep the hour in good months: the time is most at risk when the problems have just come down.
Perlow, “The Time Famine: Toward a Sociology of Work Time”7Field study: Nine months with one 17-person software team in one US firm in the 1990s. Productivity was self-rated on a 5-point scale, there was no control group, and the researcher’s presence may have sustained the policy. Selected pages read.A quiet time policy, blocks without interruptions three mornings a week, raised the share of engineers rating their productivity above average to 59% and then 65%. When the researcher returned months later, little of the effect had lasted and the engineers had gone back to old patterns; a later attempt by managers to bring quiet time back failed. Perlow concludes that such changes do not last unless the underlying work-time habits, such as a crisis mentality and rewards for individual heroics, change too.Protected time can fade once nobody keeps it.Proposed: name one person who keeps the hour and check each month whether it ran.

Health care: problems worked around, and the cost of protected learning time

Study or sourceSetting, design and sampleKey findingWhy it mattersHow to Apply It to Your Derm Clinic
Tucker and Edmondson, “Why Hospitals Don’t Learn from Failures”8Observational study: 26 nurses at nine US hospitals, observed for 239 hours, with data collected around 1999-2000 and qualitative coding of 194 observed failures. Nurses only. Pages 55-64 read.For 93% of problems, nurses did what it took to continue care without addressing the cause; only 7% of responses met lenient criteria for root-cause problem solving. The authors tie this partly to staffing designed without slack time.Without time set aside, busy clinical staff work around a failure instead of fixing it.Proposed: use the fix minutes for one lasting repair, after any patient at risk has been looked after the same day.
Tucker and Spear, “Operational failures and interruptions in hospital nursing”9Observational study: 11 nurses observed in six US hospitals, with surveys in 48 units; published 2006. Abstract only.Nurses met an average of 8.4 work-system failures per 8-hour shift; the five most frequent types (medications, orders, supplies, staffing and equipment) made up 6.4 of them, with eight mid-task interruptions per shift. The authors suggest improvement processes triggered by failures, a different design from a fixed monthly slot.Failures are frequent enough that a team always has a candidate for its one fix.Proposed: handle the urgent part the day it happens, then bring the cause to the fix minutes.
Mertens and colleagues, “Workplace learning through collaboration in primary healthcare: a BEME realist review”10Realist review: 42 publications from 6,930 references on workplace learning in primary care, mainly GPs and nurses, 1990-2015; mostly qualitative and mixed studies. Abstract only. The review does not test protected time as an intervention with measured outcomes.The reviewers recommend that policy makers and managers make sure professionals have protected time set aside for learning.Health-care reviewers point to protected time as a condition for learning at work.Proposed: earmark the practice minutes in the schedule, not in spare moments.
Sholl and colleagues, “Balancing health care education and patient care in the UK workplace: a realist synthesis”11Realist synthesis: UK literature from 1998 on balancing education with patient care, across professions. Abstract only; outcome measures are not given in the abstract.Protected learning time was among the most common interventions, and funding from the organization, managed workloads and support were the mechanisms most often found to make interventions work.Protected time works when someone funds it and manages the workload around it.Proposed: plan the patient schedule around the hour and pay staff for it.
Bartlett and Bullock, “Protected learning time in community pharmacy and possibilities for upscaling”12Qualitative evaluation: A Welsh pilot that funded 12-15 days of protected learning time for community pharmacists, studied with diaries from 20 pharmacists, 15 interviews, 12 employer surveys and 3 education leads (30 participants); self-report. Abstract only (2026).Participants reported benefits for development, wellbeing and services, but employers faced higher costs for locum cover and disrupted services. Participants proposed pre-scheduled slots or non-patient-facing hours.Protected time has real costs for the business and for patient access, which is why it needs planning.Proposed: choose the slot with the least patient impact, decided before booking opens.

Rehearsal and simulation: what practice can do

Study or sourceSetting, design and sampleKey findingWhy it mattersHow to Apply It to Your Derm Clinic
Arthur and colleagues, “Effectiveness of training in organizations: a meta-analysis of design and evaluation features”13Meta-analysis: Organizational training studies across many settings, not health-care specific; published 2003. Abstract only.Sample-weighted mean effect sizes were about 0.60 to 0.63 across reaction, learning, behavior and results criteria, which the authors describe as medium to large; for behavior, d = 0.62 (122 studies, 15,627 people).On average, organizational training changes what people do on the job.Proposed: give practice the largest share of the team hour.
Cook and colleagues, “Technology-enhanced simulation for health professions education”14Systematic review and meta-analysis: 609 studies of 35,226 health-profession trainees, searched through May 2011: 137 randomized, 67 nonrandomized two-group and 405 single-group pre-post studies, with large heterogeneity. It compared technology-enhanced simulation with no intervention, so it applies to low-tech role-play only by analogy. Abstract only.Pooled effect sizes were 1.09 for process skills and 0.50 for direct effects on patients.Simulation practice improved health professionals’ skills, with smaller effects on patients.Proposed: rehearse at the real chair, with the real supplies.
LaVelle and McLaughlin, “Simulation-Based Education Improves Patient Safety in Ambulatory Care”15Program report: In-situ simulation of chest pain and anaphylaxis across 21 primary or specialty clinics and five urgent-care clinics in one US health system (AHRQ, 2008); the program authors’ own evaluation, with no control group. Abstract only.Staff gained knowledge, confidence and skills, and the drills uncovered 40 safety concerns that were then addressed.Rehearsing a rare emergency in the real clinic can also expose fixable process problems.Proposed: send anything the walk-through exposes to the fix minutes.
Copaescu and colleagues, “Simulation-based education to improve management of refractory anaphylaxis in an allergy clinic”16Small uncontrolled studies: Three small single-site outpatient studies from Canada and the US, 2019-2026, read as abstracts: pre-post self-report or qualitative interviews, with no patient-outcome measures. This row also covers the endoscopy and hysteroscopy studies cited in the finding.An allergy-clinic anaphylaxis simulation (17 participants) improved perceived management and revealed process gaps that were then resolved. An outpatient endoscopy airway program reported knowledge and confidence gains at 6 weeks and 3 months,17 and outpatient hysteroscopy scenarios, including a vasovagal reaction, improved reported confidence among 12 participants.18In outpatient procedure settings, rehearsing rare complications is feasible and often reveals gaps.Proposed: walk through a vasovagal episode at the procedure chair, using the clinic’s written protocol.

Across all eighteen sources:

  • None tested a fixed split of team time, in an airline or a clinic, and none measured protected team time in a dermatology clinic.
  • The 40-30-30 split rests on one unnamed manager’s undated statement in a teaching case, in an undefined unit; the same case gives a second, different training figure, and no other primary statement of the rule was found.
  • SIA’s accounts describe practices, not their effects; nothing here shows that the split, or the training, caused the airline’s results.
  • SIA’s figures are its own and not externally assured. Training hours rise and fall, and Scoot changed its counting method in the latest year.
  • Employee-side counterevidence: average-to-low pay, union disputes, five-year renewable crew contracts and about 2,000 involuntary job losses in the pandemic year.
  • Protected time costs money and patient access, and it fades unless someone keeps it (the pharmacy pilot, the software team and DuPont).
  • The health-care studies are small, mostly uncontrolled or qualitative, and several were read only as abstracts; none estimates an effect on dermatology patients, safety events or profit.

Principles of Excellence: related evidence from other settings

The main text explains Singapore Airlines’ practice in plain English. These sources describe it or study related ideas elsewhere; none measured what the split did for SIA.

Point in the briefRelated evidence
Every kind of getting-better work gets a share.One manager’s description in a teaching case (undated; unit undefined).1 Improvement time loses to the day’s work unless protected (management research; model and field examples).6
People get the biggest share.Executive quotation and historical course length (teaching case).1 Training hours reported for FY2025/26, not externally assured and varying by year (company report).3 Training and simulation improve skills on average (meta-analyses, abstracts only).13, 14
Every idea is evaluated.Staff Ideas in Action, 2007-2008 (qualitative case study; company data as reported).2
The time has a fixed place on the calendar.SPARK Time for Scoot office staff (company report; no start date, uptake or outcome data).3 Protected effort eroded after success, and quiet time faded once nobody kept it (management research; field study).6, 7

Our clinic suggestions: proposed, not tested

The five practices and today’s action are our suggestions for a dermatology practice, not a tested program. The clinic scenes are illustrations, not documented cases.

SuggestionWhat it draws on
Set the split before the schedule opens: the practice manager and lead physician block the hour, about 24, 18 and 18 minutes; a patient who needs urgent care is seen at once, and the practice manager rebooks the hour within the month.The manager’s 40-30-30 description (teaching case).1 Protected time works through funding and workload management, and it has real costs (realist synthesis; qualitative evaluation; abstracts only).11, 12 The minute split is the research package’s own untested proposal.
Practice gets the biggest share: the team walks through the clinic’s written fainting protocol at the procedure chair, led by the people that protocol names.Ambulatory in-situ simulation programs report confidence and knowledge gains and found fixable gaps (program report and small uncontrolled studies, abstracts only).15, 16 Simulation research supports role-play only by analogy (meta-analysis).14
One lasting repair: unrouted pathology reports escalated, reviewed and patients told the same day under the clinic’s results protocol; the required field fixed in the team hour with the sign-off that protocol requires.Work-around versus root-cause responses (observational study).8 Failure-triggered improvement (observational study, abstract only).9 Departmental continuous improvement (qualitative case study).2
Answer every idea: the scheduler’s texting idea logged and answered; any clinical instruction approved by the clinician the clinic’s protocol names.Staff Ideas in Action evaluated every idea in 2007-2008 (qualitative case study).2
Move it, never cancel it: urgent care seen at once; only the practice manager rebooks the hour, within the month; it runs in good months too, as paid time.SPARK Time (company report).3 Protected effort eroded after success (management research).6 Quiet time faded once nobody kept it (field study, one team).7
One action today: block next month’s paid team hour, write its three parts and name who may move it.No study tested this one-day step; it applies the same pattern.

Sources

  1. Wirtz J, Heracleous L. Singapore Airlines: Managing Human Resources for Cost-Effective Service Excellence. Teaching case (Case 11). © 2016, revised 2021. jochenwirtz.com
  2. Heracleous L, Wirtz J. Singapore Airlines: Achieving Sustainable Advantage Through Mastering Paradox. J Appl Behav Sci. 2014;50(2):150-170. doi:10.1177/0021886314522323. Author-hosted PDF (nus.edu.sg)
  3. Singapore Airlines. Sustainability Report FY2025/26. Published 25 June 2026. SGX filing (sgx.com)
  4. Singapore Airlines. Annual Report FY2020/21 (financial year ended 31 March 2021). 2021. singaporeair.com
  5. Singapore Airlines. Cabin Crew careers page (Singapore). Undated page; accessed 10 November 2026. singaporeair.com
  6. Repenning NP, Sterman JD. Nobody Ever Gets Credit for Fixing Problems that Never Happened: Creating and Sustaining Process Improvement. Calif Manage Rev. 2001;43(4):64-88. Author-hosted PDF (mit.edu)
  7. Perlow LA. The time famine: toward a sociology of work time. Adm Sci Q. 1999;44(1):57-81. mit.edu
  8. Tucker AL, Edmondson AC. Why hospitals don’t learn from failures: organizational and psychological dynamics that inhibit system change. Calif Manage Rev. 2003;45(2):55-72. utsouthwestern.edu
  9. Tucker AL, Spear SJ. Operational failures and interruptions in hospital nursing. Health Serv Res. 2006;41(3 Pt 1):643-662. doi:10.1111/j.1475-6773.2006.00502.x. PubMed
  10. Mertens F, de Groot E, Meijer L, et al. Workplace learning through collaboration in primary healthcare: a BEME realist review. BEME Guide No. 46. Med Teach. 2018;40(2):117-134. PubMed
  11. Sholl S, Ajjawi R, Allbutt H, et al. Balancing health care education and patient care in the UK workplace: a realist synthesis. Med Educ. 2017;51(8):787-801. PubMed
  12. Bartlett S, Bullock A. Protected learning time in community pharmacy and possibilities for upscaling: an exploratory study in Wales, UK. J Pharm Policy Pract. 2026;19(1):2629063. PubMed
  13. Arthur W Jr, Bennett W Jr, Edens PS, Bell ST. Effectiveness of training in organizations: a meta-analysis of design and evaluation features. J Appl Psychol. 2003;88(2):234-245. doi:10.1037/0021-9010.88.2.234. PubMed
  14. Cook DA, Hatala R, Brydges R, et al. Technology-enhanced simulation for health professions education: a systematic review and meta-analysis. JAMA. 2011;306(9):978-988. doi:10.1001/jama.2011.1234. PubMed
  15. LaVelle BA, McLaughlin JJ. Simulation-based education improves patient safety in ambulatory care. In: Advances in Patient Safety: New Directions and Alternative Approaches, Vol. 3. Agency for Healthcare Research and Quality; 2008. PubMed
  16. Copaescu AM, Graham F, Nadon N, et al. Simulation-based education to improve management of refractory anaphylaxis in an allergy clinic. Allergy Asthma Clin Immunol. 2023;19(1):9. PubMed
  17. Covington CC, Muckler VC, Sheldon L, Alexander R, Morgan B. Improving emergency airway knowledge and self-efficacy levels of outpatient gastroenterology staff via implementation of online education and in situ simulation. Gastroenterol Nurs. 2019;42(3):242-250. PubMed
  18. Warshafsky C, Singh SS, Garber AB. In-situ simulation for enhancing safety in outpatient hysteroscopy: development and evaluation of a crisis resource management-based training package. MedEdPORTAL. 2026;22:11604. PubMed