Sunday, April 13, 2014

Two Trends Worth Mentioning...

I usually shy away from predictions, but I think two new trends I am seeing more of recently are worth mentioning: post-materialism culture and self-navigating quadro/multi-copters.


Since about three or four years ago, I've been seeing a slowly strengthening trickle of articles and personal stories on topics like the "tyranny of stuff" and "paying for experiences instead of things". The first well constructed expression of this trend I read was Bruce Sterling's "Last Viridian Note". While such a change in lifestyle in the past was driven by more external reasoning like "save the earth", this discourse has a more personal flavor to it. Up until the Great Recession (and a little after), there was in the U.S.A., and certain other emerging and developed economies, a relentless cultural drive to have luxurious things and, often, more than one of a particular luxurious thing. Doing so signaled high status.

Today, however, in some sub-cultures of the West, owning almost nothing but a few exceedingly high quality items and developing the complete freedom of time and wealth to have endless interesting experiences holds the highest status. This shift is fascinating, because with a little skill, one needs much less wealth to create such a life, vs. the cost of a life of accumulating and maintaining a large collection of possessions. In an era of stagnant real income, eliminating the ongoing costs of possessions is becoming ever more attractive. Experiences happen, often make us happier, and then leave our lives. Unless we suffer some injury or ailment, the costs of an experience stop when it stops. People are starting to notice the "time cost" of possessions. Everything you have must, at some point, be maintained or curated. We lose that time forever. Between the reduction in available real income and the time spent on curation, shifting to a life of interesting experiences that make us happier leaves us with both more money and more time.


Much of the startup chatter today is about disrupting this or that. Usually, either the disruption is of mundane things, or the basic business math does not really hold up in the long run, or the disruption is entirely within our online lives ("We're going to bring 'social' to ordering fast food online"). Actual "in real life" disruption, like horse and buggy to automobile, does not come around too often because of the great costs involved in developing a new technology. Two such technologies are emerging and merging, quadro/multi-copters and self-navigation. I think they will change our daily experience of transportation before 2020. In my understanding, while multi-copters have more rotor units, the whole system is simpler to manage and easier to fix.

Like some past deep shifts in technology, multi-copter technology started in universities and the toy industry. Toys have notoriously razor-thin profit margins, so getting the complex flight behavior, efficiency, and reliability of a multi-copter into a profitable toy-priced package bodes well for future scaling. Toy multi-copters have already been imbued with self-navigation and self-organization behaviors, driven by sharply falling costs of GPS technology, model-based design, and the ongoing concurrent trends of miniaturization and power reduction for computers. This means that a particularly difficult aspect of a new technology, the mathematical models running the technology, are already simple enough and mature enough to sell toys profitably.

Consider a multi-copter harness around a single standard shipping container or around a locked together block of containers. This would enable air delivery of the products inside with much less airport infrastructure - especially if the flight is fully automated. Multi-copters need similarly small infrastructure to helicopters, but their software model driven multitude of direct-driven rotors can have much better recovery characteristics than helicopters in the case of single rotor unit failure. Then consider, with enough safety engineering, the equivalent of an automated aerial train system without the need for the dedication of large land tracts to airports.  It would take less infrastructure to build out such a system than an equivalent rail system. Such an infrastructure is certainly a strong candidate for enabling people mobility in rural areas of the world with limited rail infrastructure, like Africa and Siberia.

Saturday, February 1, 2014

Lean - A Primer

A Brief Primer on LEAN

An living document of my overall understanding of LEAN. This document is licensed under Creative Commons License
LEAN - A Primer by Adam S. Keck is licensed under a Creative Commons Attribution-ShareAlike 4.0 International License.
Based on a work at
Permissions beyond the scope of this license may be available at

Based on:

Four Capabilities of Accelerating organizations

  • Specify precise approach to every task with built-in tests to detect failure and stop on error.
  • Swarm to solve problems – immediate RCA.
  • Share new knowledge throughout organization.
  • Leaders lead by developing above capabilities in every single employee.

Work Flow and System Design

  • LEAN is a framework for successfully designing complex systems and work flows. A system is a collection of related work flows. A company can be viewed as a single complex system that  provides value to customers in return for money.
  • LEAN is a way of working: Every employee uses LEAN work flow every day to both follow and improve the processes for which they are responsible.
  • LEAN is a specific implementation of Deming’s “Plan, Do, Check, Act” (PDCA)
  • Work flows should stop on error or self-regulate (“autonomation”): Each step has built in tests for verification.

On failure, stop the process and trigger an immediate RCA (Autonomation)

  • Start from the delivery of the correct work flow output to its consumer. Each downstream need paces and specifies work upstream in the process (Kanban is a specific method of achieving this goal).
  • Develop the work flow by working backward from the output that exactly fulfills the needs and requirements of final customer (i.e., what Raving Fans calls the Ideal).
  • Mistakes: Human error generally considered only when a person does not follow the current written process or does not verify that the output of each step is correct. All other errors are considered defects in the process that allow errors to occur. This philosophy drives the “Swarm to RCA each failure” capability.

Work Flow Creation Framework (In order)

  • Specify outputs – What does the work flow have to deliver, to whom, when, and what does it mean for the work flow to be successful?
  • Design pathways – Flow of materials, information, and services. Who is specifically responsible for each step in a pathway?
  • Design step connections – Linkages between adjacent process steps
  • Specify task methods – How exactly is each step in each process accomplished successfully?

Work Flow Creation Tools

  • Checklist
  • Automation code
  • Flow chart
  • Input/Output/Handoff chart

Problem Solving (Iterate)


  • Defect-free work flow
  • On-demand work flow
  • Work flow provides only exact output needed by client process or customer.
  • Immediate fulfillment of needed output.
  • Work flow runs without waste
  • Work flow is safe and secure (personnel not harmed, and information and assets secure) 


  • The commonly cited “A3 process” is a specific work flow and presentation format that implements the elements below in a way that ensures customer buy-in at each step.
  • Use graphical elements to efficiently present and confirm information with customers.


  • Background: Why is this problem important?
  • Current condition: Measurements and metrics. Get and confirm information directly.
  • Gap analysis: How do the process and its outputs differ from the ideal (see previous section)
  • Root Cause Analysis: Swarm on failure; analyze gaps from ideal
  • Develop Countermeasures (rapid prototyping = experiments to find right solutions.)
  • Specify target condition: Desired new process with countermeasures in place
  • Measure actual outcome of new process: repeat measurements and metrics
  • Gap analysis and further RCA

Sharing Knowledge

  • Organization-wide sharing accelerates productivity. Everyone follows documented work flows. 
  • See one, Show one, Do one (from hospital LEAN efforts).
  • Codify discoveries for wide dissemination (e.g. Toyota “Lesson learned books” specify what’s feasible/cost-effective for a certain type of output or process)
  • Knowledge-base that stores documented work flows = “company memory”


  • Regularly practice system design and problem solving
  • Develop LEAN skills in all employees, at every level.
  • Practice following work flows with verification of each step to prevent defects.


  • Everyone works using LEAN principles every day. Bottom up – sometimes guided - work flow improvement.
  • Everyone who knows the skills above leads those who are new by teaching them the above skills.
  • Learn to see and solve problems with rapid prototype iterations. Practice this skill.
  • Work flow improvers get and confirm information directly. Nothing is assumed.