Introduction to LEAN 4.0

Lean manufacturing is ready for its next iteration.  While Lean theory draws upon practices that have been around for centuries, the commonly practiced version of Lean today is somewhere between 60 and 80 years old dating back to the early days of TPS.  The market, however, has changed quite a bit since the 1940s and 1950s.  This blog does not intend to question the core value of Lean as an operating philosophy; generally speaking, Lean in its current form works and works across a broad spectrum of environments.  Rather, we believe that today’s landscape offers practitioners new capabilities, consistent with traditional Lean theory, that can take business performance and manufacturing theory to the next level.  It starts with questioning the uses of TAKT and Heijunka.

A quick introductory story

I had a client once, a medium sized industrial equipment manufacturer, that reached out to help address a stagnant growth situation.  They were a fairly mature Lean shop with robust planning and operational practices.  However, they had only managed to achieve 1% YOY revenue growth over the last 3 years despite a healthy market and growing customer base.

They would conduct demand planning analysis on a quarterly basis (every 13 weeks) to determine their line rates for the quarter and were pretty rigid about communicating product lead times to the market.  Their typical lead times were 2-6 weeks, where 2 weeks was approximately industry standard, and lead times were adjusted weekly based on their firm backlog.

Whenever large order volumes came in, the plant would smooth out the demand over future weeks and extend lead times accordingly.  Not surprisingly, as the plant whittled away at the order book, few new orders came in until lead times returned closer to 2 weeks.  The analysis concluded that they were missing ~50% of potential business due to this accordion pattern.  It’s the classic case of a self-fulfilling prophecy that all too many manufacturers deal with today.  You could grow your business if you eliminated self-imposed constraints and listened more closely to the market.

The problem with averages

“The assembly line can only work only if you have a standardized product” – Henry Ford

While Ford’s claim specifically refers to a production line, it’s reasonable to extrapolate his point to apply to a broader notion of “progressive flow.”  Furthermore, to what lengths can we extrapolate the notion of a “standardized product”?  Is it limited to product platforms, the whole portfolio, the whole product strategy?  Our contention is that it’s most useful to think in terms of the entire product strategy, e.g., design, feature, cost, lead times, etc.  Now ask yourself: how much functional standardization do I actually have across my product portfolio and manufacturing strategy?

It’s not a stretch to restate Ford’s claim as such: “Progressive flow is difficult, and potentially impossible, if you have high degrees of variation in product design and customer requirements.”  More often than not, the problems we see in manufacturing stem from businesses unsuccessfully fighting to reconcile this statement on a daily basis.  Establishing a progressive flow environment relies on planning around averages.  High degrees of variation widen the gaps between the averages, which are the basis of your production plan, and your daily production requirements.  The result is a production environment that is inconsistent with market expectations.

Quantum TAKT and Heijunka

If we’re learning anything from the current environment, it should be that customer-centricity and speed win in today’s market.  Traditional TAKT and Heijunka scheduling are inherently at odds with both.  You can imagine an environment 50 years ago where you had a handful of SKUs and you were the only player offering such products.  Given that you had your customer base hostage to a small catalog and varying lead times, you could trust averages and apply production smoothing mechanisms to level out daily requirements and keep costs as low as possible.

The game has changed, however, and customers are not tolerating anything short of how they want it, when they want it.  Ignoring this need can be a death sentence for businesses and stubbornly applying old fashioned TAKT and Heijunka practices makes you incompatible with your customers.  The answer lies in shrinking the time horizon – going from a 13-week TAKT calculation to a daily or less TAKT calculation (i.e., Quantum TAKT).  Anticipating demand in 13-week increments is not good enough anymore.  Demand in today’s environment needs to be dealt with at the lowest and most finite possible level.

Introducing Lean 4.0

All innovation in the digital manufacturing environment can boil down to improving the precision of your operations, either through asset/human capital performance or operational intelligence.  How you think about Lean’s role in the future should be no different.

In a series of upcoming blogs, we will take a more comprehensive look at Lean 4.0, but starting with this core premise is critical: if what you’re doing is incompatible with the needs of the market than it’s time to evolve or someone else will.  It’s about building in complete flexibility and agility to achieve your full market entitlement.  In a sense, it’s about eliminating (potentially all) demand planning to liberate your business to win the series of possible games versus the single game you are anticipating.  Only then will you break the shackles of a production philosophy that pulls you back to your original estimate instead of reaching your full market potential.

Step one in this journey is to set table stakes on what the market requires and work backward to optimize internal flexibility and agility.  Does the market require 1,000 variations of your core product at 2-week lead times across the portfolio?  If so, you will probably need to focus on establishing TAKT in finite increments, employing Heijunka only within those intervals, better localizing your supply chain, and taking advantage of on-demand labor to eliminate labor related capacity constraints.  Perhaps IoT, additive manufacturing technologies, and wearables would help address the perceived challenges in doing these things.  Leading companies are already orienting their strategies accordingly – today it’s a matter of playing offense but soon it will be a matter of defense.

Stay tuned for more in our Lean 4.0 series.  For more information about Veryable on-demand labor for manufacturing and warehouse applications, please view our website at: www.yourlaborpool.com or www.veryableops.com

Digital Manufacturing: Onwards to 2018

If you were waiting all year for the big digital manufacturing revolution and subsequent manufacturing renaissance, don’t worry you didn’t miss it.  While it’s true that U.S. manufacturing output and employment have risen significantly over the last year, this is more the product of overall economic health and less the result of any wide adoption of a new manufacturing ethos.

New paradigm shifts in manufacturing are rare and driving change in manufacturing takes a good bit of time.  But whether it’s 5 years from now, 10 years, or 20 years, digital manufacturing is where we are heading and you should be thinking about what you’re doing to prepare your business for it.  Adopting digital manufacturing techniques today is playing offense.  If you wait too long, you’ll be playing defense.

The Digital Manufacturing Framework

In a previous blog series, we laid out some basic frameworks for how to think about the broader digital manufacturing use case.  In this blog, we will overlay a couple of these frameworks to orient use cases evolving in today’s landscape.

Any time you hear about an emerging manufacturing technology, it will likely fall into one or more of four general categories of use cases:

— IoT / Connected Factory – operational intelligence, big data, analytics, etc.
— Automation – robots, cobots, factory control, etc.
— Additive Manufacturing – 3D printing, etc.
— Workforce Enablement – on-demand labor, augmented reality, haptics and optics, etc.

In addition, each of these use case categories derives value by impacting what we call four manufacturing focus areas:

— Machines – predictive maintenance, OEE, asset productivity, etc.
— Materials – inventory optimization, materials availability, etc.
— Flow / Flexibility – bottleneck optimization, real-time production monitoring, etc.
— Labor – labor productivity, production flexibility and agility, etc.

It’s important to consider both of these dimensions when evaluating the digital manufacturing landscape since only through the two combined can you build a business case that links technology to use case to value creation driver to value created.

Where do we stand at the end of 2017?

With few exceptions, we approach the end of 2017 in the same place we were a year ago.  This should come as no surprise to most – manufacturing as a sector is stubborn and conservative.  It’s difficult to drive meaningful and lasting change, and doing so requires the coordination among many stakeholders, inside and outside the four walls of the factory, as well as solution design, implementation planning, and the allocation of capital.

So we stand today with an idea of what the future looks like and a general understanding of the technologies and use cases that enable the future vision.  The following table gives some examples of those use cases.

The fact of the matter is that manufacturing change takes time and the value creation mechanism has to be clear.  Implicitly, everyone knows that each of these use cases could create tremendous value but the level of uncertainty around implementation requirements, time to value, vendor selection, etc. has led to another year of general stagnation.

What should we expect in 2018?

Until there are tangible case studies and demonstrated value in the marketplace, expect the digital manufacturing revolution to continue to make only modest gains.  From our vantage point, the pace is being stalled by a preoccupation with the most difficult use cases: those involving machine intelligence and central IT infrastructure.  It’s an odd place to start due to the difficultly of implementation but it’s also no surprise that this is the case.  The providers of machine-oriented IoT based solutions like GE, Siemens, Rockwell Automation, etc. are the same OEMs with an existing installed base and have been first to market solutions.  The large tech companies with cloud based operational intelligence platforms are also in the game but lack the installed base at the equipment level.

The fastest way to drive large scale adoption is to refocus on the primary sources of value like labor productivity and growth enablement.  At Veryable, we believe that labor flexibility should be the tip of the spear that subsidizes other value add use cases as illustrated in the figure below.

For most manufacturers, the most productive way to think of machine and flow oriented use cases is to view them as enablers for higher labor productivity and throughput.  Starting with labor will put the business case in your hands and the rest will follow logically.

For more information about Veryable on-demand labor for manufacturing and warehouse applications, please view our website at: www.yourlaborpool.com or www.veryableops.com

Veryable and The Third Wave

The Internet of Things is coming to a theatre near you.  Some of the early developments will be subtle, e.g., new apps popping up for your phone, new technologies arriving in your grocery store, new services springing up in your neighborhood, etc.  However, expect to turn back 10-15 years from now and find the 2017 landscape as unrecognizable as you might view 2002, before smart phones and social media became such permanent and essential fixtures in your life.  What you will be observing are the gradual effects of what Steve Case refers to as the “Third Wave of the Internet.”

The Three Waves

In his 2016 book called The Third Wave, AOL co-founder Steve Case outlines what he calls the three waves of the internet; the first two having already passed and the third closely on the horizon.  Case characterizes the First Wave of the internet as essentially the infrastructure phase.  This phase began in the 1980s as companies like AOL, Microsoft, Cisco, etc. established the infrastructure and connections to create the internet itself.  He characterizes the Second Wave as the building on top of the internet: new technologies to access information and connect people.  This started to gain steam around the turn the century and continues in high form today with notable examples like Facebook and Google.  Most of the big tech companies of today are the major players of the Second Wave.

The Third Wave, however, is shaping up to be a different animal – one with some similarities to the first two waves, perhaps most akin to the first, but unique in its invasiveness into traditional sectors and everyday activities.  The banner under which this wave resides is often called The Internet of Things.  Simply put, we’re talking about connected items, edge technology, operational intelligence platforms, predictive capabilities, autonomation, and machine learning.  These capabilities and technologies have a wide range of use cases that span nearly every sector and end market.  Thus, conversations about IoT often meander down a confusing tunnel of considerations and implications.  The common thread is that IoT is entering into some unchartered territory in familiar and often stubborn environments.

IoT will Impact Every Sector

The scope of the Third Wave pays homage to the First Wave in that there is a massive infrastructure element and culture change requirement.  For the most part the infrastructure components are ready, as evidenced by the avalanche of sensor technologies into the market, the proliferation of IoT platforms, and rapid advances in edge technology.  The challenge, however, is that the solution looks a bit different depending on the sector, environment, and application, i.e., the architecture varies on a case by case basis.  Effective IoT solutions will therefore require a significant amount of industry knowledge and access.

For the Second Wave, in particular, Silicon Valley has been a hub of innovation.  New products and services are incubated out West and distributed through PCs and mobile devices.  This is a perfectly reasonable model for Second Wave solutions that target individual consumers or business enterprise IT applications.  In contrast, Third Wave solutions demand transformation.  Transformation cannot be driven remotely with packaged solutions – it requires knowledge of the current paradigm coupled with an understanding of the “art of the possible” and the capabilities to deliver the result.  With 75% of Fortune 500 companies residing in states that receive less than 25% of total Venture Capital funding, the landscape is primed for new entrepreneurs to deliver these solutions within their markets.  Steve Case calls this the “Rise of the Rest.”

The Challenge of the Third Wave

The primary implications of the Third Wave are underscored by a recent IndustryWeek article by Steve Minter that aptly contends that the “toughest challenges of IoT are not the technology.”  If you consider an oil field services company, for example, looking to implement smart assets in the oilfield and achieve higher operational productivity, an appropriate IoT solution begins with the understanding of the current work flow and the access needed to implement the transformation.  The challenges are not in the technology per se, rather the solution architecture, environmental considerations, work flow changes, program management, skills, and training.

At Veryable, we believe that the answers to today’s manufacturing challenges require solutions born in our manufacturing heartland; solutions that specifically address the needs of the sector.  In a previous blog titled New Demands on Manufacturing, we talk about some of these challenges such as higher demand variation, lower lead time requirements, levels of customization, etc.  What manufacturers need today is higher levels of flexibility and agility with zero cost to scale enabled by IoT based operational intelligence.  Robots and additive manufacturing will be game changing components of the future digital manufacturing environment but basic blocking and tackling need to be mastered before we introduce the trick plays.  The combination of IoT technologies and on-demand labor will do just this.  Major players in the Third Wave will understand these types of needs because they understand the complexities of the sectors they serve and the behaviors within.

For more information about the on-demand labor for manufacturing and warehouse applications, please view our website: www.veryableops.com

Digital Manufacturing and Labor

The large-scale redistribution of operational intelligence is one of the most understated consequences of digital manufacturing and other IoT based operational capabilities.  In the previous blog called Digital Manufacturing, we explored how to think about Digital Manufacturing as a specific category of use cases and this installment will pick up specifically on the labor implications of such use cases.

Building the Business Case

Any emerging digital manufacturing solution is going to require significant capital investment and implementation costs.  While the costs will vary dramatically based on the solution provider and the solution architecture, the cash outlay will always precede time to value.  In fact, many businesses will be challenged to meet 1 or 2-year payback hurdles and thus required to maintain a longer-term manufacturing strategy vision.  Therefore, getting the business case right upfront is critical for all manufacturing companies.

While most operations leaders intuitively grasp the digital manufacturing value creation potential immediately, quantifying the business case can be a challenge.  Because new IoT based capabilities impact all manufacturing metrics and most value drivers, business cases must be able to monetize the benefit of higher absorption, improved quality, better safety compliance, increased throughput, better inventory visibility, and increased labor productivity – that is, the value creation potential in the business case has to be the sum of these incremental operational improvements.  In the eyes of many senior executives and finance types, this is a tougher sell than a direct cost out project with a more straightforward benefit calculation.

Operational Improvement Benefits

While all five manufacturing metrics – productivity, inventory, service, quality, and safety – stand to improve drastically through digital manufacturing, the direct impact on the P&L is trickier to quantify.  At the end of the day, the direct P&L impact is going to lie with asset productivity, cost of quality, and labor productivity.  The first two are inherently linked into the digital manufacturing ecosystem: capabilities like machine-to-machine communication and predictive maintenance are directly embedded in the solutions that are coming to market now like GE Predix and Rockwell Automation’s Connected Enterprise.

However, the labor productivity aspect is more of an indirect benefit and viewed as a byproduct of the overall productivity enhancements.  Realizing the labor productivity benefit is going to be harder than most companies realize; this a real problem since it is commonly the highest addressable cost bucket and the biggest driver of the business case.

Labor Productivity in the Digital Manufacturing Environment

Let’s start with a simple thought experiment by considering how labor productivity is realized under the following digital manufacturing scenarios:

a.  New machine capabilities increase equipment uptime and throughput

b.  Plant wide material flow visibility anticipates where bottlenecks are about to form in the plant

c.  Machine learning capabilities automatically level load the production schedule in real time

d.  RIF based track and trace solutions monitor the movements of employees and material

e.  Smart cameras identify value add versus non-value add work

f.  Digital work instructions, wearables, and operator alerts show workers what to do in real time

These example use cases all point to some level of improved labor productivity but the quantified benefit is less than obvious.  For example, higher machine throughput will improve absorption and potentially decrease labor cost per unit and digital work instructions should add an element of operator enablement, and thus individual productivity, as it relates to facilitating process steps and identifying the right material and tools to use.  The key takeaway here though is that, under the current thinking, labor productivity is a lagging benefit – one that comes to fruition as a byproduct of these use cases rather than the primary focus area.

A Better Approach

Without addressing the primary constraint with today’s labor paradigm, a digital manufacturing solution will always be sub-optimized.  There are two primary weaknesses:

1.  The imbalance between asset flexibility and labor flexibility

2. The lack of direct labor benefit in the digital manufacturing business case

The flexibility imbalance is really where the system falls apart.  In the ecosystem of the future where production planning, customer service, and plant equipment can all respond and adapt in real time, labor becomes a prohibitive bottleneck.  Labor has to respond with a complimentary level of flexibility and agility to enable the system to achieve its full potential.  Couple that with the lack of direct labor benefit in the business case and we have an obvious obstacle.

At Veryable, we believe that the answer lies with harnessing the power of the redistribution of operational intelligence to enable a fully variable and on-demand labor model.  A piece-work based on demand labor capability is the key that fully unlocks the digital manufacturing explosion within the next 5-10 years.  With an on-demand labor model, companies can build in productivity from Day 1 and establish essentially an infinite amount of flexible capacity to match the new capabilities embedded in the rest of the solution architecture.

Rather than implementing digital work instructions and trying to figure out to derive a labor benefit, start with the labor benefit available from an on-demand model and figure out how to enable an untrained resource to be fully productive using only these instructions.  Rather than using cameras to identify non-value added time, use recorded information to identify quality issues in real time or illustrate process steps in front of on-demand resources.  In short, harness the new intellectual property and tailor it to enable anyone to be productive.

We believe this to be the number one opportunity for digital manufacturing adoption and success, as well as the manufacturing sector at large.  Can you imagine being able to flex on a day’s notice to double or triple your labor capacity?  Can you imagine an economy where workers’ job security is not tied to one specific job?  Can you imagine an economy where every worker is relevant, and a prospective resource, for every manufacturing or distribution company?  We can.

For more information about the on-demand labor for manufacturing and warehouse applications, please view our website: www.veryableops.com

Digital Manufacturing

The last few years may have introduced more new capabilities into the manufacturing sector than most manufacturing practitioners will see in a lifetime.  Many industry experts expect, as a result, that manufacturing will change more in the next five years than it has over the last 20 years.  The change is being driven by a suite of new technology innovation from the Internet of Things to robotics to wearables.  We refer to the use cases for this suite of enabling technology as Digital Manufacturing.

What is Digital Manufacturing?

In a previous blog called Megatrends: Future Impact on Operations, we explored the implications of technological breakthroughs on manufacturing labor.  Here we will dig into what these breakthroughs are and how they are impacting the industrial environment.

Undoubtedly you have heard a recent increase in the use of terms such as Internet of Things, the industrial internet, Industry 4.0, or simply digitization.  These are all being thrown around to describe the overall emerging technology being adopted by companies to enhance product portfolios, provide new or better services, and/or improve internal operations.  Buried within this landscape are specific capabilities that are converging to form the Digital Manufacturing category of use cases.  As shown in the figure below, the spectrum of Digital Manufacturing elements can be broken down into four groups from broadest implications to narrowest: Connected Factory, Automation and Robotics, Additive Manufacturing, and Man-Machine Interaction.

In theory, a complete Digital Manufacturing solution would include aspects of all four of these elements, and many companies will be heading down this path within the next five years.

Which elements are having the biggest impact?

While much of the recent discussion out of Silicon Valley has focused on the implications of robotics, our experience and analysis points to Connected Factory as the biggest game changer.  The Connected Factory essentially represents a controlled application of the Internet of Things within the four walls of the plant or across a network of plants.  With the emergence of cheaper sensing technology, data storage, track and trace capabilities, and data analytics, sites can have the ability to establish operational intelligence platforms that house and analyze all aspects of the manufacturing process in real time.  The days of disparate and disconnected specialized systems (e.g., MES, WMS, CRM, etc.) are numbered.  While these systems may still serve a purpose in the future architecture, they will merely be feeders into the operational intelligence backbone.

The number of specific use cases and value creation opportunities stemming from the Connected Factory are endless.  With this comes the ability for operations managers to have real-time visibility to material flow, worker movements, inventory transactions, and machine performance.  In addition, the machine learning capabilities will be able to reprioritize production sequences, send alerts to operators, digitally deploy work instructions, and proactively identify potential bottlenecks or quality issues before they arise without management intervention.  These use cases all stem from the operational intelligence within the Connected Factory.

What is the value creation opportunity?

The Connected Factory is ushering in something unprecedented in the history of manufacturing: a large-scale redistribution of operational intelligence.  Even in most of today’s production environments, much of the operational intelligence still resides with the individual operator.  This is why training and structured onboarding processes are still high priorities for most businesses.  The inefficiency with this historical paradigm is that individual operators only have a small amount of influence over the overall site operations, i.e., they can only optimize the variables that are within their control.

A recent study of ours concluded that the amount of non-value added labor activity at most sites is greater than 50% of total labor cost.  Only a small portion of the non-value added cost is typically related to anything performance based – the real efficiency barrier is the inability to coordinate all the variables in the production environment to keep operators working efficiently.  More often than not, they have to devote valuable time to find material, locate tools, re-sequence work orders, or consult with engineering.  The Connected Factory directly addresses this overall coordination of all of these variables.

Now back to the redistribution of operational intelligence.  Along with the capability to coordinate activities, the operational intelligence backbone becomes a massive repository for company intellectual property.  It contains all product and production information, measures and analyzes performance, and deploys the information as designed.  The implication is that harnessing and deploying this IP appropriately will enable a future state where individual experience and expertise are redundant, and potentially even counter-productive.  This will have an enormous impact on how we manage manufacturing labor in the near future.

We will explore the labor implications, specifically, in the next blog in the Digital Manufacturing series.

For more information about the on-demand labor for manufacturing and warehouse applications, please view our website: www.veryableops.com