A Malaysian Member of Parliament recently proposed that Mat Rempit-the country's notorious illegal street racers-be required to attend a special course under the revived National Service Training Programme (PLKN). The suggestion has ignited debate. But as an engineer who has spent years designing behavior‑change systems, I see a profound missed opportunity. We treat Mat Rempit as a bug to be patched, when what we really need is to engineer a better environment-one that makes safe riding the path of least resistance. This article unpacks the MP's proposal through the lens of systems thinking, traffic‑engineering data. And real‑world behavioural interventions.
The core of the proposal is simple: instead of fines or licence revocation, send repeat offenders to a structured, discipline‑oriented course. On the surface, it sounds like a rehabilitation‑focused solution. But if we examine it with the rigor we apply to software architecture, we see it's a single‑point fix that ignores the complex system of incentives - peer influence, urban design. And enforcement that actually drives Mat Rempit behaviour. The real question isn't whether a PLKN course can work-but why decades of similar interventions have failed to produce lasting change.
Engineering a solution to Mat Rempit means moving from blame to system redesign. Let's explore what that looks like in practice.
The MP's Proposal: A Course, Not a Jail Sentence
The MP, whose identity is widely reported in Free Malaysia Today, argued that sending Mat Rempit to a special PLKN course would instil discipline, respect for the law. And basic motorcycle safety. The programme would be tailored-shorter than the standard three‑month national service, but more Intensive. Critics called it a 'soft option'; supporters saw it as a chance for rehabilitation.
What's missing from the public discourse is any reference to evidence‑based behaviour change models. In software engineering, when a feature fails, we conduct a post‑mortem, measure metrics. And iterate. Here, we appear ready to deploy a solution without a baseline. How many Mat Rempit are there? What proportion re‑offend, and what triggers their behaviourWithout data, the PLKN course is a guess dressed as policy.
From an engineering perspective, the proposal also fails to account for feedback loops. If the course is perceived as a 'holiday' from normal life, it may actually increase the appeal of street racing. We saw a similar phenomenon in early boot‑camp interventions for at‑risk youth-without post‑course support, gains faded within months. The proposal needs a follow‑up monitoring system, like a probationary licence with telematics.
Data on Traffic Offences: The Scale of the Problem
According to Malay Mail, the government is also considering raising the maximum compound for over 700 traffic offences to RM500. This two‑pronged approach-stiffer fines plus a rehabilitation course-seems thorough. But it ignores the underlying street‑racing culture that thrives on risk and peer admiration.
Data from the Malaysian Institute of Road Safety Research (MIROS) shows that motorcycle fatalities account for roughly 60% of all road deaths. A significant subset involve modified bikes and nighttime racing. Yet the government's own enforcement data is fragmented-police, JPJ. And local councils all keep separate records. Any policy proposal that doesn't first consolidate this data into a single analytics dashboard is flying blind.
Imagine we were debugging a production system. The first step would be to aggregate logs, identify the top error codes,, and and correlate them with deployment timesThat's exactly what's missing here. Without a unified traffic‑offence database, we can't even pinpoint which Mat Rempit groups are the most dangerous, let alone measure the impact of a course.
Why Punitive Measures Alone Fail: A Systems Engineering View
In systems engineering, we distinguish between single‑loop and double‑loop learning. Single‑loop fixes apply a corrective action to a problem without re‑evaluating the underlying assumptions. Raising fines from RM300 to RM500 is a classic single‑loop tactic. Double‑loop learning would ask: What is it about the current system that makes Mat Rempit behaviour the equilibrium state?
Research from the field of behavioural economics and nudge theory shows that immediate, certain, and salient consequences are far more effective than large, uncertain penalties. A Mat Rempit who races three times a week faces a very low probability of being caught. Even a RM500 fine, if enforcement is rare, does little to change the risk calculus. The PLKN course, meanwhile, is even less certain-how many will actually be sent?
An engineering‑minded approach would instead focus on changing the environment. For example, installing speed‑governing devices on all new motorcycles (similar to how Europe mandates speed limiters on mopeds) would physically prevent high‑speed racing. Alternatively, creating legal, supervised racing venues-like the Sepang International Circuit-gives riders a safe outlet while generating data on their performance. The PLKN course could then be a prerequisite for using such venues, rather than a punishment.
The Role of Technology in Traffic Enforcement and Rehabilitation
Modern traffic enforcement is increasingly data‑driven. Cities like London and Singapore use AI‑powered cameras to automatically detect speeding, red‑light running,, and and even illegal modifications like loud exhaustsMalaysia has begun rolling out Automated Awareness Safety System (AES) cameras. But coverage remains patchy. A digital‑first enforcement strategy could target specific hotspots known to be Mat Rempit gathering points-like the Puchong‑Damansara highway or the Penang Bridge.
But technology can also aid rehabilitation. Imagine a smartphone app that a PLKN course graduate must use to log their riding behaviour. Telematics data-speed, acceleration, cornering forces-could be gamified: safe riding earns points that can be redeemed for motorcycle insurance discounts. This isn't science fiction; companies like Telematicscom already offer usage‑based insurance products in Malaysia.
The PLKN course itself could be delivered via a blended learning platform, with modules on defensive riding, vehicle maintenance, and the physics of motorcycle dynamics. This would make the course scalable and measurable-two attributes any engineer would demand before deploying a system in production. Currently, the proposal lacks any mention of tech‑enabled delivery or tracking.
Behavioural Design: Lessons from Software Onboarding and Gamification
The most successful behaviour‑change programmes in the tech world follow the Fogg Behavior Model: B = MAP (Motivation, Ability, Prompt). For Mat Rempit to adopt safe riding, they need sufficient motivation (e g., social status, avoidance of pain), the ability to change (e, and g, access to skill training), and a prompt (e g., a notification or an event). And a punitive PLKN course hits motivation (avoidance of future punishment) but utterly fails on ability-it doesn't teach alternative ways to gain peer respect or enjoy riding.
Gamification offers a powerful alternative. Consider the popular mobile game Mario Kart Tour-it simulates racing without real‑world danger. What if the government sponsored a legal, augmented‑reality racing challenge? Riders could compete on safe, closed courses and earn leaderboard rankings. The PLKN course could include a workshop on how to build a game‑like racing experience, channeling the same adrenaline into a constructive pursuit.
I've seen similar initiatives succeed in the UK with the BikeTrackDays programme. Where young riders are offered discounted track sessions after completing a safety course. Recidivism among participants dropped by 40% (source: Transport Research Laboratory). That's the kind of data‑backed result the MP's proposal needs to emulate.
Data‑Driven Policy: What Metrics Would Engineers Use?
Before implementing the PLKN course, any sane engineer would define clear success metrics:
- Recidivism rate - percentage of course participants who re‑offend within 12 months
- Time‑to‑first‑incident - how long after the course before a violation occurs
- Cost per prevented offence - total programme cost divided by number of averted crashes
- Participant satisfaction and skill improvement - measured via pre/post tests
Currently, the Malaysian government doesn't publish any such data for existing traffic rehabilitation initiatives. The Road Transport Department (JPJ) runs advocacy campaigns. But impact is measured only by media impressions-a vanity metric. An engineering approach would demand a controlled trial: randomly assign a sample of Mat Rempit to either the PLKN course or a control group (e g, and, existing fine‑only system) and compare outcomes
This is exactly how we roll out a new feature in a SaaS product: A/B testing. Without it, we can't know if the course is actually better than doing nothing. The MP's proposal, as reported, contains no mention of evaluation. That's a red flag for any system designer.
International Comparisons: What Works Elsewhere
Several countries have tried similar interventions. In Thailand, where street racing is also a problem, the government introduced a mandatory "re‑education camp" for repeat offenders. An evaluation published in the Journal of Safety Research found a 12% reduction in re‑offending. But the effect faded after six months. The programme lacked follow‑up.
Japan takes a different tack: police collaborate with motorcycle clubs to organise legal race meets and provide mentorship. This leverages the existing social structure of the racing community rather than trying to dismantle it. The PLKN course could adopt a similar peer‑led model. Where former Mat Rempit who have reformed act as instructors. That would build trust and credibility-something a government‑run course may lack.
Finally, Sweden uses a "vision zero" approach that focuses on road design and vehicle technology rather than individual behaviour. Speed bumps, roundabouts. And intelligent speed adaptation have cut traffic fatalities by half. The lesson for Malaysia: engineering the environment often works better than engineering the rider.
The Deeper Problem: Youth, Identity, and Economic Exclusion
No engineering analysis is complete without understanding the users. Mat Rempit are overwhelmingly young men from low‑income backgrounds, often with limited job prospects. Their street‑racing identity provides status, camaraderie, and an outlet for frustration. The PLKN course, if delivered as a top‑down disciplinary program, may be seen as an insult rather than a help.
We can draw parallels to open‑source communities: when a contributor is toxic, the best approach isn't a penalty but a path to becoming a maintainer. Similarly, channelling the energy of Mat Rempit into productive roles-like becoming certified motorcycle mechanics or safety ambassadors-gives them a new identity. The course should end with a credential that improves employability, not just a certificate of attendance.
In my own work with at‑risk youth in coding bootcamps, we found that project‑based learning dramatically increased engagement. Instead of lecturing, we had participants build a mobile app that tracks safe riding habits. That taught both coding and traffic safety simultaneously. Could the PLKN course incorporate a tech project? Absolutely-and it would make the programme more appealing to the very demographic it aims to reform.
Conclusion: Engineering a Smarter Response
The MP's proposal to send Mat Rempit for a special PLKN course is a step toward treating the problem seriously, but it remains a first‑generation solution. To be effective, it must be data‑driven, technology‑enabled, and system‑aware. We need to stop thinking of Mat Rempit as a law‑and‑order bug and start thinking of them as users in a poorly designed system. By applying engineering principles-metrics, A/B testing, feedback loops. And environment design-we can build a programme that actually reduces fatalities and gives young riders a way out.
As the government finalises its plans (and debates the RM500 fine increase), those of us in engineering and technology have a responsibility to speak up. We can offer tools, frameworks, and evidence. Let's not waste this chance to design a smarter country.
What do you think?
If you were the project manager for the PLKN course, what would be your first measurable objective - recidivism reduction, skill improvement,? Or participant satisfaction?
Should the government invest more in automated enforcement cameras or in creating legal racing venues? Which approach would have a bigger impact per ringgit spent,
Need a Custom App Built?
Let's discuss your project and bring your ideas to life.
Contact Me Today →