015-0013

An Integrative Thinking Approach to Organizational Learning

Dr Pauline Found, CardiffUniversity, Lean Enterprise Research Centre, Cardiff, CF24 4AY, United Kingdom. Tel: +44(0)29 2064 7022. Email:

Robert Kearney, GKN Aerospace, PO Box 500, Golf Course Lane, Filton BS34 9AU, United Kingdom Tel: +44(0)117 317 5000. Email:

POMS 21st Annual Conference

Vancouver, Canada

May 7 to May 10, 2010

Abstract

This paper investigates the problem of what affects the learning process. The nature of learning is defined as a “Mess”, where many factors interrelate. Integrative thinking is chosen as the method to investigate this Mess. The literature of organizational learning and the learning organization is reviewed, as well as the differing definitions of what constitutes learning. The concepts of system thinking and system dynamics are then used to determine causality between factors that affect learning. The paper reports the findings of recent research which concludes that the senior managers have the most effect on employee learning, until peer pressure outweighs this. This is governed by a formula that can be applied to determine the relationship.

Introduction

There is a desire to create a workforce that improves itself of its own free will; one that does not need projects imposed on them, and audits of previously imposed solutions to maintain them. The question is how do we achieve this? Some of the problems with conventional management thinking such as poor system, and poor behavior of both managers and workers exacerbate the problem. This paper will attempt to investigatethe main issues, and propose some way forward for the future to achieve an engaged and truly ‘Lean’ environment. A principle that comes to mind is that people are not responsible for their behavior; the system is (Deming, 1986). How then, can a system be designed so that people want to improve? Is this even possible? Is the system the correct thing to focus on?

This paper attempts to unravel the problem of achieving Continuous Improvement (CI). We begin first by looking at Lean and CI’s role within it. We then move on to look at what sort of problem achieving CI is, and how we might best think about this. System dynamics and the learning processes are then covered, which lead to a literature survey on factors that may prevent learning. These factors are investigated as the part of the research and the results are analyzed and presented. Finally conclusions and suggestions for further work are proposed.

Relationship to Existing Literature

Continuous Improvement (CI) is being investigated within the context of a Lean manufacturing environment. Lean is the formal label created in the 1980s by Krafcik (Krafcik, 1988) and popularized by Womack and Jones to describe their findings of how Japanese manufactures were significantly outperforming western companies. Lean was defined as having five principles (Womack & Jones, 1991):

  1. Specify Value from the customer’s perspective
  2. Identify the Value Stream
  3. Make the value Flow
  4. At the Pull of the customer
  5. Strive for Perfection

The Lean Enterprise Institute (LEI) defines the core idea of Lean as being:

“..to maximize value whilst minimizing waste. A Lean organization understands customer value and focuses their key processes to continually meet those needs.”

The use of the word continually and the fifth principle of Lean, striving for perfection;indicates that Continuous Improvement (CI) is central to this definition of Lean as companies change and adapt to changing customer requirements and market needs.

A law of ecology states that for an organism to survive, its learning must be equal to or greater than the rate of change of its environment (Revans, 1982):

L ≥ C

The rate of change of the business environment is governed by incremental changes, as existing products are developed; and more drastic shifts, as entirely new products enter the market. It would appear then, that it would be beneficial to look at how to create learning as a whole, and not specify incremental or breakthrough activities (or Continuous Improvement and Continuous Innovation).

This changing market is a result of a more knowledge driven economy:

“Productivity and competitiveness are, by and large, a function of knowledge generation and information processing: firms and territories are organized in networks of production, management and distribution; the core economic activities are global – that is they have the capacity to work as a unit in real time, or chosen time, on a planetary scale.” (Castells, 2001:52)

Learning is key to maintaining a competitive advantage within this changing knowledge based economy. It is worth pointing out that a recent book from the LEI is entitled “Managing to Learn” (Shook, 2008). The key point of this is that the A3 report drives a method of thinking that ‘reframes all activities as learning activities at every level of the organization’.This highlights the role of learning within a Lean organization.

Achieving learning within an organization can be considered a problem that needs to be solved. All problems are not of the same type. This section looks at different problems types and a thinking method that can be applied to them. Learning is defined as one of these problems types and a method of thought to investigate the problem is selected.

Before designing a solution to a problem, the problem itself must be framed and specified (Lockard, 2000).

Problems have been separated out into two types (Rittel & Webber, 1973):

  • Tame problems
  • Wicked problems

The concept of a wicked problem was originally developed in the social planning sphere (Ritchety, 2008). But the concept has been applied to strategic planning (Camillus, 2008). Tame problems possess the following characteristics (Conklin, 2001):

  1. They have a well-defined and stable problem statement.
  2. They have a definite stopping point.
  3. They have a solution that can be objectively evaluated as being right or wrong.
  4. They belong to a class of similar problems that can be solved in a similar manner.
  5. They have solutions that can be tried and abandoned.

Wicked problems, on the other hand, have the following characteristics (Rittel & Webber, 1973):

  1. There is no definite formulation of a wicked problem.
  2. Wicked problems have no stopping rules.
  3. Solutions to wicked problems are not true or false, but better or worse.
  4. There is no immediate and ultimate test of a solution to a wicked problem.
  5. Every solution to a wicked problem is a “one shot operation”; because there is no opportunity to learn by trial-and-error, every attempt counts significantly.
  6. Wicked problems do not have an exhaustively describable set of potential solutions, nor is there a well-described set of permissible operations that may be incorporated into the plan.
  7. Every wicked problem is essentially unique.
  8. Every wicked problem can be a considered to be a symptom of another wicked problem.
  9. The causes of a wicked problem can be explained in numerous ways. The choice of explanation determines the nature of the problems resolution.
  10. The planner has no right to be wrong.

A wicked problem may possess some, or all, of these ten characteristics (Camillus, 2008). Wicked problems are also defined as being divergent (Hancock, 2004). A tame problem is convergent, in that the more it is studied; the more solutions eventually converge. Divergent problems do not promise solutions – the more they are studied, the more differing solutions are presented.

An alternative definition of a problem is a Mess (Ackoff, 1974, Horn, 2001). A Mess is a set of interdependent problems that are identifiable only by their interaction. The solution needs to be designed to be able to respond to the Mess. The solution cannot be the aggregation of independently obtained solutions to the parts. It needs to be dealt with as a whole systematically. (Ackoff, 1981).

These competing definitions can be drawn together to give the following problem types:

Figure 2 Matrix of problems types

This can be compared to the Cynefin framework (Snowden & Boone, 2007)

This gives us to four problem types:

  • Tame – convergent problem. - simple
  • Mess – convergent ‘Wicked’ problem. - complicated
  • Wicked – divergent problem. - complex
  • Black Holes – a divergent tame problem. – Chaotic. A black hole exists if the same actions are applied repeatedly eventhough there is evidence that they are not giving the desired effect.

The nature of the problem will determine how the problem can be solved. Tame problems can be solved by a linear process of data collection, analysis, formulation of a solution and implementation (Hancock, 2005). Messes and Wicked Problems, on the other hand, need to be understood by modeling. The problem cannot be broken down into distinct components that are then solved and the component bolted together to create a solution (Ackoff, 1981).

So what sort of problem is creating organizational learning? The problem must be convergent as others have achieved it, the most obvious example being Toyota, in that employees work to improve their process (Spear & Bowen, 1999, Liker, 2004). However, there are many relationships between different factors that make it complex, or a Mess. A method is required to help build understanding of this complexity and how to untangle it.

Since Taylor and the scientific method (Taylor, 1947), attempts have been made to see problems in different lights and therefore reach different and better solutions than previously thought achievable. It is now possible to design a way of thinking based on how the mind actually works which will help to avoid a mindset of judgment, which although useful, is not enough to deal with complex problems (de Bono, 2009). How a problem is thought about provides the highest leverage for developing a better solution within both manufacturing (Shook, 2008) and services (Seddon, 2005).

Integrative thinking (Martin, 2007) is emerging as a framework for combining together a number of different tools, techniques and ideas into a whole thought process, which lends itself to looking at Messes and Wicked Probelms. This framework is based on the four-stage model shown in Figure 3.

Figure 3The Integrative thinking model (Martin, 2008)

Each stage is described in more detail.

1. Salience

Salience is the first step. Information that is deemed to be relevant to the problem is determined. Perception is responsible for over 90% of errors in thinking (de Bono, 2000). It is this step which attempts to negate this by taking into consideration more salient information than would normally be the case. The concept of mental models is used to achieve this. Humans do not perceive reality, but a mental model of reality that is composed of the information that the individual finds relevant (Sterman, 2003). Each individual will create a different mental model. This difference in models creates conflict. It is normal to advocate one’s own model to prove the other incorrect. Integrative thinking recognises that all models are wrong, and the conflict of models merely shows more salient information from other models that need to be considered. This is supported by Mitroff, who states:

“A single view or perspective of any problem is automatically wrong. It cannot hope to capture all of the subtleties and the complexities that are characteristics of real problems.” (Mitroff, 2004, citied in Martin, 2008)

2. Causality

Using tools such as Casual Loop Diagrams (CLD) (Senge, 1990) and System Dynamics (Forrester, 1961), causal links between salient information are constructed. This helps to deal with the complexity of the model and also helps ensures a lasting solution is developed.

3. Architecture

With causality between salient information determined, a solution is designed. The concept of design is becoming an increasingly common theme among business thinking (Martin, 2007) (Seddon, 2003) (de Bono, 2000) (Neumeier, 2008) and is set against the traditional judgmental method of architecture (de Bono, 2009). The design process is key to this stage of working (Brown, 2008). The design process can be loosely described as a three-stage process:

  1. Inspiration
  2. Ideasation
  3. Implementation

The process often feels chaotic to those used to linear, milestone based processes which are typical in most modern organizations (Brown, 2008). A project may loop back to previous stages as prototyping and learning occur.

4. Resolution

Once the designed solution is implemented further salient and causal information may become evident and these problems are feed back into the solution to refine its design.

It is interesting to point out the similarities to the methodology used in Managing to Learn (Shook, 2008), which used a three-step process of understanding causality, seeking predictability and reflection to ensure learning.

Based on the literature, we assume that continuous improvement (CI) is a Mess, and not a Wicked problem as CI has been achieved in other areas, Toyota being the obvious example. As the problem is a Mess, some method of understanding it is required. Integrative thinking has been selected as this method. The study reviewed the literature to identify salient points that may affect learning. In a second step, causality is examined using the concept of system dynamics. A dynamic model of learning is found with two feedback loops, one for learning a new skill and one for forgetting this new skill. The salient points were added to each of these feedback loops. The model shows there is a tipping point in the learning process.

Figure 4. The learning process

A number of factors may affect the learning process. Organizational defensive routines which protect individuals, groups and organizations from embarrassment or threat (Arygris, 1990), (Argyris, 2002). This prevents learning. It tends to be the case that the recommendations made to negate these routines reinforce the causes of them. Companies tend to make two mistakes in their efforts to become a learning organization:

1) Learning is defined too narrowly as problem solving. This focuses the learning experience on the external environment. It is argued that if learning is to be successful, people must look inwards; to reflect critically on their own behavior and identify ways they inadvertently affect the organizations problems.

2) Assuming getting people to learn is largely a matter of motivation.

This means organizations tend to focus on:

  • New organization structures
  • Compensation programs
  • Performance reviews
  • Corporate cultures

Double loop learning however, is not a reflection of how people feel (i.e. are they motivated or not), it is a function of how they think, and how they think is made up of the cognitive rules people use to design and implement their behavior. To avoid these problems companies must make the ways managers, and employees, reason about their behavior a focus of organizational learning and CI programs. Therefore a key factor in preventing learning is not employees’ attitudes about change or commitment to CI, but the way they reason their behavior and that of others. A theory of action that controls behavior is created which is designed to consistently modify behavior according to four basic values:

  1. To remain in unilateral control
  2. To maximize winning and minimize losing
  3. Suppress negative feelings
  4. To be as rational as possible.

The purpose of these is to avoid embarrassment or threat, and causes people to keep private the premises, inferences and conclusions that shape their behavior and avoid testing them in an objective fashion. By not sharing their theory of actions, people’s behavior is not affected, and this causes them to revert to type when moved into a new role, further preventing learning (Garratt, 1990).

Focusing on an individual’s attitudes or commitment is never enough to produce real change.Even when people are genuinely committed to improving their performance and management has changed its structure in order to encourage the right behavior, people still remain locked in defensive reasoning.

To prevent this, the first step is for managers to examine their own behaviors critically and change their own theories of action. Unless this occurs any change activity will be short lived. Change has to begin at the top otherwise defensive senior managers are likely to disown any transformation in reasoning patterns coming from below. This indicates that a perceived lack of learning at the bottom of an organization may be due to a lack of learning at the top.

The role of managers is further developed by Lawrence & Losh(1967), where it is stated that their behavior is heavily influenced by their cognitive emotional orientation. This is comprised of time, interpersonal and the quality and degree of the organizations structure’s formality.

Further investigating the cognitive aspects of learning, it is argued that the mindset that promotes efficient execution inhibits employee’s ability to learn and innovate by preventing experimentation and reflection on mistakes (Edmondson, 2008). This work identifies four practices, which will form the basis of a learning infrastructure:

  1. Using best knowledge to design specific process guidelines.
  2. Enabling employees to collaborate in real time.
  3. Routinely capturing process data to discover how things are done.
  4. Reflecting in an effort to find ways to improve.

The work goes on to state four points that negatively affect learning:

  1. Critical information and ideas fail to rise to the top – people become reluctant to take up managers’ time with anything other than certain and positive information. They do not offer concerns, ideas, or question what they are being told. This links to employee performance reviews, as no one wants to admit mistakes – the environment becomes one of survival, rather than one of learning (Seddon, 2005).
  2. People don’t have enough time to learn – switching to a new approach can lower performance in the short run. Managers that overemphasize efficiency discourage employees from adopting new approaches.
  3. Unhealthy internal competition arises - high performing individuals or plants are often rewarded. This can make people reluctant to share ideas or best practices. An idea is to develop absolute rather than relative performance incentives.
  4. Companies think they can do no wrong - if a company is successful whilst inhibiting learning, then the management team will associate efficiency with high performance and therefore drive more of the same behaviors.

To institutionalize learning, the environment must be made safe. Organizations must foster psychological safety. Toyota is given as a company that has enabled this by constantly encouraging employees to find problems. Psychological safety is crucial if employees are to collaborate and make decisions without management supervision, which are classic components of a Lean method of management. (Edmondson, 2008)