IT operations can learn from lean manufacturing
This article has been podcast
Keep an eye on Lean. It is the next big thing (fad or real change?). I always watch what is coming across from manufacturing to IT because - in the service management area at least - that is the trend: manufacturing teaches us.
So I was interested in one paragraph in a recent McKinsey article
...as the lean approach percolates into ever wider circles of operations, it ceases to be about best practice and starts to become a part of the fabric of doing business.
which is likewise what we all aspire to when introducing ITSM.
Operating executives in many sectors are adopting lean techniques rapidly, so soon they will no longer be a differentiating factor; the important thing, in the heat of competition, will be how well companies implement them.
This is the most important point for me. Best practice means looking like everyone else. If you want to be competitive you either need to (a) do it better, as in the article or (b) do something else. I've had an article planned on that thought for ages.
This next level of the lean journey is managing the softer side of the equation—less about tools and frameworks, more about building the energy and engagement of employees from the shop floor and the office pool upward, tapping into their ideas, focusing them on constant problem solving, and keeping them open to change and flexibility
i.e. People -> Process -> Things
The IT Swami says Lean will be a hot topic of ITSM conferences in the next year or two. It is not in the index of the new Service Strategy. I don't have my Service Operation here but I bet it's not in that either. But it will pop up.
What interests me here is the repetition of common themes. We can learn as much from manufacturing's experiences as from their ideas.
Made in New Zealand 
Comments
Where is Lean
Had nor seen this discussion before.
Looks like Lean did not become the new big thing?
Aale
PS First comment from iPad. Bought this from UK as they are not yet available home. I'm still resisting iPhone.
Look before you Lean
Go back and take a closer look at the Service Strategy book. You will see ideas and concepts which are analogues to product engineering and production engineering. The idea of outcomes (utility) and certainty of outcomes (warranty) have their roots in operations management. If you understand the principles behind Six Sigma and Lean, take another look at Fig 3.3 and 3.4. (Think control charts, process capability, etc). The idea of objectives, resources and constraints are from the field of industrial engineering and operations research. You will find a diagram which depicts how strategy produces a system of constraints (Fig 7.10) which is graphical representation of a set of simultaneous equations. You solve it to obtain a set of feasible solutions. This principle has been used in manufacturing for decades. The whole idea of core service package is an analogue of a product platform or chassis in manufacturing. Anybody spot an obscure reference to "OEM segment"? Guess what the "M" stands for in Fig 5.27? This is much like unbranded FirstData processing the credit card transactions of a regional bank. What about the Taguchi Loss Function? (Fig 9.1). Where do you think that concept has come from? Take a closer look at Fig 7.8. If you have spent anytime in production engineering and operations management, you will see know the predecessor of workflow is the assembly chart. The Kano Model is a product engineering classic long before Lean became a fad in the services industry. Take a closer look at Fig 7.13 to Fig 7.15. That stuff is based on lean manufacturing principles. Instead of an assembly line or manufacturing cell, think of data centers, call centers, or as the Service Strategy book calls it "Service Units". Instead of hosting product lines as in a factory, you host services and service contracts. All the three-box diagrams with +ve and -ve arrows are loosely based on the concepts of systems dynamics and control engineering.
Anyway, I think we should "look before they lean". :-)
P.S. This discussion forum is good ... congratulations to the owner
Lean and ITIL
Lean production (a term coined by John Krafcik at MIT, now the VP of strategic planning at Hyundai) originates from Toyota’s TPS. The father of TPS, Taiichi Ohno, had a habit of greeting new engineers by drawing a chalk circle on the factory floor. He would tell them to stand there and look for waste until he returned. Later in the day, Mr. Ohno would ask the engineer what he had learned. Mr. Ohno would subsequently inform the recruit he was a moron, pondering out loud why Toyota ever hired such an idiot. This exercise would be repeated until the hapless recruit finally figured it out. (answer at the end)
Too often, companies get excited about a new initiative and charge off. They get great initial results and the concept withers and dies, becoming yet another failed management fad. I've observed many organizations engaged in kaizen, initially removing waste, but then having no clue what to do next.
When automobile manufacturers decided to embrace TQM and emulate Toyota, they copied its factory floor practices. They installed pull-cords to stop the assembly line, they used statistical control charts to detect variations in quality, they practiced just-in-time inventory to lessen inventory.
Yet all still trail Toyota in quality, design features and productivity. Why? The reason is that organizations copy the most visible and obvious practices. Unfortunately, these are often the least important practices. The key to Toyota’s success is not from a set of techniques but from its philosophy. Instead of copying what they do, manufacturers should have copied how Toyota thinks.
When we examine the thinking behind Lean Manufacturing and ITILv3, we can see that Lean principles are indeed reflected within ITILv3 principles. For example:
- Lean manufacturing refers to an evolving dynamic process of production spanning the total enterprise. Its model emphasizes an "evolutionary process of change and adaptation, not an idealized technology-driven end state." (ref. the ITIL Service Lifecycle).
- At the core of Lean is the determination of value; defined as something for which a customer is willing to pay. All other aspects of the manufacturing process are deemed waste. Indeed, one of the hurdles to Lean is understanding the idea of waste. In Lean Manufacturing the term waste refers not to physical materials but rather the relationship of the resource to the end customer. In other words, if the customer won't pay you for it, it is waste. (ref. Marketing mindset, Utility and Warranty, Service definition)
- The traditional view in manufacturing (and services) is: Cost + Margin = Sales Price. Lean says this is incorrect; this model incorrectly reflects actual practice. Rather, Competition and customer demand will set selling prices. The Lean view is instead: Profit = Sales Price – Cost. (ITIL Financial Management, Return on Value)
- Lean relationships between suppliers and primary producers hinge upon collaboration rather than command and control. For example, collaborative risk-sharing, cost-sharing and information-sharing arrangements. The primary technique for visualization and diagnosis is Value Stream Mapping (VSM); a means for understanding process interactions, stocks, flows and delays. VSM, like the Parolini Method, is understood to be a Value Network tool for supply chains and manufacturing. Indeed, VSM modelling and simulation is a hot topic in “proving” the benefits of Lean transformation projects. (Value Networks/System Dynamics)
- Tradition manufacturing uses “batch and queue” systems, all designed for mass production. Similar processes are grouped together in a chained linear sequence. They depend on economies of scale to lower their costs, thereby requiring large inventories. No one needed to remind Toyota that Japan was a nation with few natural resources. Further, there were many intrinsic problems to this approach. Such as what Toyota called the Forrester effect, aka the bullwhip effect. Or the inability to detect defects until further downstream. Or the cost of delay and latency; a can of soda that takes 30 minutes to drink may sit in a warehouse for 30 days. Lean instead implements “single-piece-flow”. This is only accomplished by dealing with the system as a whole; not independent components. Hence Ohno's circle - he wanted the engineer to take-in the dynamics of the factory floor taking place all around him, not just in the circle. (Systems Thinking)
There are more but I've belabored the point.
ERP
Fascinatingly enough, Lean’s emphasis on "evolutionary process of change and adaptation, not an idealized technology-driven end state" has caused it to come in conflict with ERP implementations.
Lean focuses on pulling demand directly from the customer, responding promptly and continually improving process. ERP, on the other hand, focuses on forward planning and the movement of data from a central control perspective, needing to track every activity and every piece of item on the factory floor.
The former has workers only performing activities that add value to the product; the latter has workers recording data and bar-coding to keep track of inventory and labor. Seasoned Lean manufacturers are increasingly skeptical of ERP, questioning the need to permanently institutionalize a wasteful process by laying an ERP system over it.
Tom Grace published the paper “ERP and Flow: The Status Quo Meets Its Replacement”, where he writes, "Flow replaces ERP on the shop floor and also cures the IT headaches that ERP inspired in factories." He explains that Lean manufacturers have come to see ERP as a liability, perpetuating inefficient workflows and unnecessary data collection.
The skepticism is echoed by analysts like AMR research, “You can spend more time bar-coding than manufacturing products”, “Companies get no shop-floor productivity out of ERP”, and Coleman Consulting Group, “ERP gives information, but it does not necessarily add to the bottom line.”
It will be interesting to see how this plays as the Lean movement grows in the services sector.
Lean is important AND...
let's be careful.
When I saw that McKinsey article hit my inbox, I dragged it down immediately. Overall, it's a good article and I was pleased to see it.
While I agree lean is important for those of us engaged in service management, I think one of the key messages to be conveyed is that it's not about lean being the next answer.
I believe that considering any approach or framework (be that lean, ITSM, Six Sigma or whatever trips your trigger) as a solution unto itself is inviting problems and hardship.
One of the underlying principles in lean is a change in orientation. I think of it as something more akin to a lifestyle alteration (moving this into a personal context).
Someone who's having health problems can keep eating the greasy, fatty, sugary (what I'd call "yummy") foods, even after the doc has told them not to and that the consequences are significant. Yet, just knowing that may not be enough to help cause the sustained alteration of lifestyle required. Something else might be required to make that happen.
Adopting lean techniques, applying them to service management as a part of the next evolution of the delivery organization is entirely appropriate. This will naturally call for an alteration (in the organizational sense).
I think it's also worthwhile to see the good practices (embodied in lean) that an organization (or at least some of its people) has already made its own. Let's give credit where it's due.
kengon
Lean in IT is happening now
Lean is spreading from manufacturing to many other fields, not just IT.
And in IT, it's not just about eliminating waste during implementation of IT projects. It even extends into software development. Consider reading Lean Software Strategies: Proven Techniques for Managers and Developers, by Peter Middleton and James Sutton. Details can be found at http://www.productivitypress.com/productdetails.cfm?PC=327.
Lean Service IS Coming...
Skep,
I couldn't agree more with your view that lean service is heading our way in the next 6-12 months.
It's only a matter of time before one of the larger consultancies recognises that to win lucrative contracts to upgrade (or implement) ITIL V3 then a pretty radical new approach is required.
The drivers for this are the ones we would expect to see: -
- CIO's are fed up of forking out for large, costly initiatives that have rather dubious payback
- Service Executives are keen to leverage new approaches - especially ones that promise efficiencies just by executing the methodology ('Lean')
- Service Managers are always (quite rightly) on the look out for anything that can take them forwards, get them the V3 'badge' and also streamline their operations
Now, having said all that, what about the real people who run and optimise service? The thousands of folks whose livelihoods might be affected by the impact of lean initiatives?
In Manufacturing, especially the Car Manufacturing industry over the last 5 years, 'lean' has meant only one thing - job losses.
Robots, automation, six-sigma results and the constant pressure to get "more done for less" has inevitably led to companies downsizing and outsourcing. This is the way of the world as companies seek to deliver higher returns to shareholders in an ever competitive industry.
However, we Service professionals have an 'edge'. At least I think we do.
We have an edge just because of the industry that we work in. That edge lies in the fact that we deliver, improve, optimise and continually improve the services we deliver. Just by virtue of the fact that the word 'service' is in our job descriptions (somewhere) means that we have some form of human contact, interaction and response.
Sure Service Desks can get outsourced to lower cost delivery countries (as we have seen) and large ITSM organisations can take over the overall running of a companies services (as we have also seen) but the days of full service automation - making the types of decisions that require rapid, but considered, decisions are still far away.
[Side Bar: We struggle to get tools to perform correctly - CMDB case in point!]
This all doesn't fall on a plate for most of us though. Just because we work in Service doesn't mean that we have the right to be somehow magically protected from the advent of 'lean'.
We all have to be prepared to "pay the price in advance" and commit to keeping our skills, knowledge and ultimately our experience levels right up to date. Those lower down the Service Chain - towards the technology only end - will suffer.
But the days of automated tooling driving out service jobs to replace humans in the services arena - to the extent that it has in manufacturing - are still a long way off.
That doesn't mean to say it won't happen - economics will determine that - but at least there's an 'air gap' of a few years before it is likely to happen and impact us.
Where 'lean' does have a value over the next two years lies in root of ITIL with Process efficiency. If we ask ourselves the question how well executed, repeatable, robust and predictable are our service management processes - then most of us would be lying if we said "VERY". In truth, the vast majority of organisations do not fully understand how the humans involved in service processes are performing, to what level and what can be done to improve them.
I will be making the case, when asked in 2 years no doubt, what should be added to ITIL Version 4.
One of many items I will be suggesting is the addition of 'time' tracking within process flows and their descriptions.
What does this mean?
Well, in the ITIL process flows there is no account for (a) delays, (b) queues, (c) step failures and (d) volume changes. All of these factors affect the timing of process flows from one step to the next.
Manufacturing have done a tremendous job (underpinned by Six-Sigma no doubt) with this area.
When I visited a large car manufacturer three years ago to learn more about how they apply six sigma to their processes - the major thing that struck me was how they could measure (in real life!) the delays that their processes were subjected too - and then how they could systematically eliminate these delays to dramatically improve the predictability of process flows.
[Side Bar: They even identified that incoming component deliveries would be delayed by traffic queues outside the plant - so they had a traffic management and coordination team that would monitor and communicate with traffic authorities to streamline incoming deliveries]
I fear far too few organisations meet these practices within Service.
I'd be interested to learn more about what other Skeptic readers think of this topic.
I also absolutely stand to be corrected if there is an onslaught of people writing in explaining how they have already integrated lean disciplines into their Service Delivery (outside of manufacturing).
So, let the 'lean' debate begin...
Lean is alive and well
HHmm.. I am going to go out on a limb here and suggest that those inside of IT, you know who you are - may have skipped reading some of the core beliefs inside of lean books, and then perhaps forgot to track the latest advances caused by the accidental slip of not thinking customer first.
There he goes again I hear some of you shout - customer, customer customer. Well you see Lean and the lifeboat I am on - outside-in (or customer centric) thinking, share a few mantras and methods - the lean 'learn to see' (the customer) and OI - map the customer experience (careful of the jagged edges in both - the do you actually know what you are doing edges) - in the form of : "give the customer exactly what they want, when they want it and how they want it". Damn those pesky customers. They always get in the way of a good process.
Lean six sigma, business process management and six sigma folks have arrived at the conclusion that process improvement (maturity?) absent the customer interest - is a total waste of time. see www.bpm-summit.com event next year. yes I know - Lean and six sigma have their customer concepts (customer satisfaction and results and voice of the customer, respectively) but they have both rather neglected actually involving the customer on too many occasions - and paid the project failure consequence.
So while the LSS/BPM folks retrace their steps, in walks the ITSM/ITILers looking for an alternate buzzsaw - "I know we will use Lean - management like lean". They ignore or forget the fact tlean has a three legged stool as I mentioned, customer sat, results and waste based upon what the customer will pay for - note thats three times I mentioned customer.
And - they try and squeeze as much waste out of ITIL like practices, or even put them, on a fat reduced diet and suggest we all implement Lean ITIL, or a leaner incident management process (WTF?). And they fail.... again.
For Pete's sake (sorry pete), take a moment to sit with your end users (customers?) and leave the ITIL books at home - and get to grips with what those who pay your salary actually want out of IT - they likely want "invisible technology" to enable them to get their important work done - without failing (ASQ where are you when I need you).
Lean, outside-in, six sigma, ITIL, ITSM, etc etc - they have one thing in common - the word 'customer'. Do me and everyone else a favor - go hug one today. Go watch what they do and the experience they encounter when using your 'IT Service' - then throw away your ITSM process toys and build a system to serve them. You will find that lean methods will help - as will six sigma, but neither will offer a blueprint for success, just a means of progressing towards what you need.
The answer is not in these methods - its in your customers.... and lean is alive and well as a set of methods to help you find what your customers want - if you respect it enough to actually read the good guidance out there and mix that into the old as dirt requirement - to serve your customers.