Pursuing Perfection Beyond the Acceptable Quality Level (AQL)

An ad promotes an Internet Service Provider’s (ISP) subscription plans.  On the bottom in small fine print is written “30% minimum speed at 80& reliability.”

The Philippines’ National Telecommunications Commission (NTC) in a memorandum in 2011 mandated that ISPs should provide at least 80% service reliability to customers: 

An ISP therefore should be able to provide an internet connection to its customers at or more than its minimum internet speed 80% of the time.  Or to put it another way, the ISP’s customers should be able to experience the minimum internet speed they signed up for 24 out of 30 days in a month.  If a subscriber does experiences the minimum speed six (6) days or faster in a month, the NTC considers the ISP’s service acceptable. 

The ISP is also required to tell its subscribers what the minimum internet speed is. In the ad mentioned above, the ISP informs customers that its minimum speed is 30% of what it advertises.  Thus, if a subscriber applies for a 300 Mbps plan, the ISP will guarantee a minimum speed of up to 90 Mbps.  An ISP is obliged to give only up to 30% of what the subscriber signs up for. 

Just imagine if this kind of advertising is applied in the fast-food industry.  A customer orders from a fast-food restaurant and the restaurant is guaranteed only to serve 30% of what’s on the menu.  If we order a 10-piece bucket of fried chicken, for example, and the fast-food gives you only up to three (3) pieces and the government allows it, we would surely be angry but we won’t be able to do anything about it. 

What the ISP advertises and what it actually serves evolves from the concept of the Acceptable Quality Level (AQL). 

The United States military adopted the Acceptable Quality Level (AQL) as a standard for inspection during the Second World War.  The idea is to set a level of what would be considered acceptable from a batch of items received from a vendor or a factory. 

For example, the US Army may accept a lot of 10,000 bullets if only up to fifty (50) of the bullets (AQL of 0.5%) are defective.  The US Army would be willing to pay for all 10,000 bullets even if it really would be able to use only 9,950 of them.  That doesn’t sound so bad unless you’re the soldier who ends up with the fifty (50) bad bullets. 

How AQLs are set varies from enterprise to enterprise, industry to industry.  Vendors would plead for higher AQLs if customers are buying items in very large lot sizes.  Sellers of small parts manufactured in large quantities like nails, wires, screws, and welding rods would ask for high AQLs as they would argue that defects are unavoidable and impossible to sort & separate all unconforming items. 

Customers, however, would discriminate what items deserve higher (looser) or lower (stringent) AQLs.  Customers would insist that how AQLs are set should depend on what the items would be used for.  For construction of a large warehouse, for instance, having a higher AQL for a large number of nails may be tolerable.  For owners of residential homes, however, they may not welcome a high AQL for nails or any construction material as this may result into a badly built house that can bring inconveniences if not hazards (e.g. a bad nail that leads to a collapsing ceiling). 

Food product manufacturers may accept higher AQLs for not-so-critical items like packaging materials but won’t welcome allowances for defective raw materials.  A food enterprise may accept a few imperfect boxes but I doubt it would accept even a few bad apples out of hundreds. 

And pharmaceutical firms won’t probably be too tolerant for high AQL’s for ingredients for medicines.

Utility firms apply AQLs in their services.  Electricity firms negotiate contracts with communities and try to convince consumers to accept a minimum level of power un-reliability, such as allowing for a number of days for a power plant to not supply electricity so that it can undergo maintenance.  Water companies try to get customers to agree on an acceptable quality of potability, to the extent notifying customers that the water they supply will not be fit for drinking anytime.    

In exchange for higher AQLs, enterprises sometimes offer discounts or defer price increases, from which they can position themselves as low-cost suppliers versus rivals.  In other words, customers can get products and services cheaply but they won’t get 100% quality. 

ISPs try to out-compete each other via this latter scenario.  Some offer the cheapest rates, advertise maximum speeds, but in the fine print guarantee only a minimum speed at a fraction of the time that is compliant to the law. 

Some ISPs will try to outdo the other such as by bumping up the minimum speed to 40% versus a rival’s 30%.  To the subscriber, however, he or she will never really get what they wish for from the advertising.

When ISPs advertise their maximum speeds, they reveal what they are capable of supplying to their subscribers.  But they also are insecure that their operations won’t run perfectly and reliably all the time.  Hence, they build in allowances to attain what they believe is an attainable level of performance which they can realistically provide to their subscribers. 

The problem with this kind of thinking is that it encourages complacency. 

The ISPs over time will won’t feel the need to improve the uptime of their broadband connectivity and to lengthen the time of their maximum service speed since they have the government’s blessing for an 80% reliability and a minimum speed that only requires notifying customers.  And they will continue to do so as long as rivals don’t try to up the game.

Many enterprises have over time accepted the AQL and said all right to accepting so many defects in the items or services they buy.  It becomes a standard that sometimes no one bothers to see if it can be improved.  If a factory is getting 95% acceptable product and its rivals are getting just the same, executives may not see the incentive to improve; after all, the factory is competitive at least for now. 

Up-and-coming competitors challenge established companies by bucking the AQL standard, by taking advantage of the complacency that take hold in established companies.  A new ISP company for instance may offer guaranteed minimum speeds of 80% (though they may lower the maximum speed advertised) at the same price as rivals.  

Competitors, the successful ones anyway, will claim to do better by offering better quality that improves from an industry’s AQL.  The real good ones would adopt continuous improvement that lead to zero defects. 

Acceptable Quality Levels (AQLs) were established to provide some reasonable standard in the inspection of items.  They weren’t meant to set the standards of quality; doing so only inspires complacency and encourages stagnancy. 

In a world where competitive disruption is more likely than ever, perfection in quality, via zero defects, is what we should pursue. 

About Overtimers Anonymous

We Need Better Monitoring Systems

Most executives like performance measures.  Otherwise known as metrics, key performance indicators (KPI’s), analytics, or scorecards, enterprises embrace performance measures as a means to assess how their businesses are doing.

The point of a performance measure is to check how an individual or team is doing against a target that is set by superiors.  (No matter what people may say, it’s always the superior who sets the targets).  Targets are set in line with strategic goals.  Individuals and teams strive to perform such that resulting measures would meet targets to attain the goals. 

But after more than twenty (20) long years since they’ve become popular, performance measures are no longer good enough, especially for supply chains. 

Supply chains are product and service streams.  Materials, merchandise, and information (printed and digital) flow through networks within and between enterprises.  From one operational step to the next, products and services transcend in value as they make their way to their final destinations: the end users.

Supply chains are sensitive to disruption.  When a disruption hits one process, every part of the supply chain feels it.  A delay in the loading of a truck, for instance, may entail a change in production schedules at a manufacturing facility it is supposed to deliver to, which in turn may cause a shortage of a product the facility is supposed to make. 

Performance measures are popular as many people could relate to them.  They are simple and easy to appreciate.  They show how a person’s work is doing versus a target that fits to that person’s tasks.  The performance target would be linked to higher levels of performance measures that would finally connect to a strategic goal. 

Unfortunately, performance measures do not work very well when there are disruptions.  Whereas they are designed such that different levels of an organisation can be made accountable for them, performance measures are not flexible to changing circumstances.  

For example, a production line supervisor is accountable for how many overtime hours his crew works in a week.  His target is that each crew member does not work more than 4 out of 40 hours of overtime per week.  He controls the overtime by rotating his crew members’ leaves such that not many of them have days off at the same time.  But if the supervisor receives a surprise rush order such that he has to make double his weekly volume, he would be forced to ask his crew to go on overtime to meet that order.  His boss, however, would ask him later to explain why he exceeded his weekly overtime target. 

Disruptions are nothing new for supply chains.  They can be big or small.  They are the results of both adversities and opportunities  And they can come periodically or frequently.  They are never identical in cause and they sometimes come in the most mundane manner, like a surprise doubling of a production order such as in the example mentioned above

Performance measures work when supply chains run routinely, much like in a game of sports.  Sports games operate under fixed sets of rules and conditions.  Players score and meet goals to win. But if it rains, the game stops.  In similar fashion, supply chain professionals perform to achieve objectives set by schedules under favourable and predictable working conditions.  But if someone changes the schedule or everyone has to go home because of a disruption like a virus-causing government-mandated lock-down, the performance measures become useless. 

Disruptions are normal.  They aren’t exceptions.  Disruptions occur often as a result of frequent adversities and opportunities that ripple through the fast-paced interconnected world we live in.

What supply chains need are monitoring systems that tell us not only what is going on but also notify us when there is a need to respond.  We need monitoring systems that will tell us about upcoming disruptions and give us time to take action.             

Two things comprise a monitoring system:   visibility and guidance.  Visibility in the form of real-time information and guidance in the form of alerts to events that merit a response. 

An example is a fuel gauge in a car.  The gauge provides visibility on how much fuel is there in a tank.  It also gives guidance via a flashing light that alerts the driver that the fuel tank is almost empty. 

Monitoring systems are not new to supply chains.  Manufacturing managers harness instruments and gauges to monitor production lines and facilitate process control.  A number of enterprises have adopted technologies such as radio-frequency identification (RFID) tags, block-chains, and artificially intelligent command-and-control systems to oversee supply chains even from long distances.   

Many enterprises, however, have had little success in mitigating disruption in their supply chains despite the growth of high-tech monitoring systems.  This is because many monitoring systems aren’t focused towards disruption.  Instead, they are geared towards performance for the sake of measuring results versus strategic goals, which as aforementioned don’t really contribute very much in a frequently disruptive environment. 

We, therefore, need to re-orient supply chains towards monitoring for disruption, not performance.  By watching out for disruption and responding to it, supply chains would be able to muster resources to mitigate it, even perhaps take advantage of it. 

One doesn’t have to start with an intricate, complicated or expensive system.  One can begin with simple reports from various operations along the chain.  For instance, vendors, brokerages firms, and shipping companies can email the status of orders for imported materials. 

Import status report

A status report such as the one above can tell stakeholders about impending issues such as a shipment that’s about to be considered abandoned and subject to penalties.

Supply chain engineers can make improvements step-by-step by tailoring feedback systems to fit different processes.  SMS texts summarising daily customer orders, entered orders in the database, communicated factory orders, MRP II real-time plans are examples.     

      

A supply chain monitoring system can also be like a tsunami warning system: 

Or it can be manifested like a dashboard for supply chain professionals to see:

Whatever the design, the purpose of the monitoring system is to allow stakeholders to watch out for disruption and respond when needed. 

Performance measures have not proven to be helpful in our disruptive-driven world.  We need monitoring systems that provide visibility and guidance especially for supply chains.  They don’t have to be complicated; they just have to be adequate enough to bring attention to disruptions.

Disruptions are a result of both adversity and opportunity.  In either case, it’s always best to be one step ahead whether it be to mitigate or to take advantage of whatever’s out there.  

Why We Need to Build Supply Chains

Enterprises are planning to rebuild their supply chains in the wake of the pandemic of 2020. 

Well, no, not really. 

Many enterprises are planning to resume production and boost inventories in the aftermath of the COVID19 pandemic.1 Some firms will narrow their product lines to those that are in high demand (e.g. toilet paper).  Others will stock up on raw materials and seek vendors that are nearer to their production sites as alternatives to risky international sources.

Not many firms, however, plan to build or rebuild supply chains.  I don’t blame them. 

Building a supply chain is not an attractive option, at least at first glance.  Most enterprises work within existing supply chains and would not outright see a good reason to build one that overlaps with other organisations. 

Enterprises would likely focus internally in their own operations if there’s any supply chain building that needs to be done.  And even then, enterprise executives would hesitate to do any major change if they perceive it would entail too much work and cost that wouldn’t reap much beneficial return.

Some companies in the past did try to rebuild their supply chains.  They called it “re-engineering” and it was popular in the 1990’s.  The idea was to redesign business operations from scratch and then apply sweeping changes to existing operations. 

It didn’t last long.2 Many companies ended up downsizing instead of changing.  A lot of people lost their jobs and companies didn’t realize much of any reward.  Re-engineering quickly lost its luster as fast as it was introduced. 

Some consultants, academics, and information technology (IT) vendors still push for re-engineering though they avoid the term.  Some pitch IT platforms such as Enterprise Resource Planning (ERP) to drive operations improvements.3 Alas, ERP and other similar platforms have not been as successful as hoped.   Many projects have fallen to the wayside

Many enterprises took to managing supply chains than into re-engineering them.  They sought talented people who’d know how to regulate inventories, negotiate with vendors, process orders, and make sure operations would comply with the latest environmental sustainability rules and occupational safety & health guidelines.  Executives also placed hopes that supply chain managers can lead in implementing new technologies such as artificial intelligence.  Supply chain management had become the norm.  Re-engineering was forgotten. 

Then the pandemic came. 

Enterprise executives know that supply chains need to change in the aftermath of COVID19.  And it isn’t just because of COVID19.  Before that, it was the tariff war between the United States and China that turned global trade upside down, not to mention similar disputes such as the British exit (Brexit) from the European Union.  There were also the supply and price fluctuations in commodities from metals, rare earths, to crude oil.  There were also the natural disasters. There were also the cyber-security data breaches.  And there were also the numerous upstart entrepreneurs who were introducing technologies such as drones, ride-sharing, video-streaming, and ecommerce mobile apps that threatened traditional businesses. 

Enterprises had to accept: supply chains, especially the global ones, were vulnerable.  They don’t work well in a disruptive environment.  The pandemic proved it.  All one had to see were the idle factories, closed warehouses, shut stores, and empty shelves.  The present supply chain set-up no longer applies. 

Building supply chains isn’t really that hard and expensive.  Sure, it requires investment but just like supply chain management, the key is talent.  One just has to employ the right people with the right talent.  The good news is that the talent is there in front of us, ready to work and available.  They are the industrial engineers; they have the tools and the skills and whom I’d rather call supply chain engineers.   

It’s not management but engineering that would drive the building of supply chains.  Engineers build things.  That’s their role.  Managers don’t do building; that’s not their role. 

The process of building supply chains is not too far off from constructing a factory or warehouse.  One has to have a plan, a timeline, and a full list of resources and costs.  And one has to have a engineer with the expertise and leadership to design and oversee. 

The difference lies in the nature and scope.  Whereas a facility such as an office and warehouse lies within the bounds of an enterprise, a supply chain encompasses the stream of products and services that crosses organisations and borders. 

A supply chain is like a river.  Build a dam and that’s a facility at a point in the river, with the purpose of harnessing the river’s water.   Building the supply chain involves setting up systems and facilities along the river to ensure the continuous and sustainable flow of water from start to finish. 

The supply chain engineer works not only with stakeholders within the enterprise but with stakeholders from other enterprises, such as but not limited to vendors and customers.  The engineer identifies and designs what needs to be built along the supply chain river. 

Most of what would be built first would likely be the networks and systems that link along the supply chain stream*.  The engineer would seek the optimal design that would synchronise and sustain flow that would uphold competitive standards of reliability, quality, and versatility. 

The 2020 pandemic is the latest in the series of 21st century disruptions to supply chains. It was the worst and it won’t be the last.  Enterprises who realize that their supply chains are vulnerable and need to be built with engineering talent would be on the right track to reviving their competitive edge.  

*Note:  I wrote a similar blog in LinkedIn in July 2019 in which I stressed structures in supply chain building.