If the Real World is so great for learning why use business simulations?
It is commonly declared that for a business simulation to deliver learning it must be real. I feel that this is a naive view as it does not explain what is meant by reality, does not take into account the problems with learning from experience in the real world, and ignores the purpose behind a using a business simulation - to provide learning that helps business people do a better job.
The chart below shows how a reality focus helps develop factual knowledge and is especially appropriate for academic education as it develops knowing. In contrast, a beyond reality focus is especially suitable to help develop business ability and through this the ability to do things better .
Reality exists in several forms - the two most important are External Validity and Psychological Fidelity.
External Validity: If you are creating a business simulation model to investigate a business problem (management science/operations research) then the model has to behave exactly like the real world situation. For example, many years ago I was involved in designing a production scheduling system. I prototyped the system and found that it did not reduce inventories as expected. Eventually, because we could not identify the flaw in the scheduling logic I created a model that simulated the material flows and the scheduling logic  and used this to explore different scheduling logics. So, external validity, involves a photo-real replica of the real world but, as explored in the section about the problems with reality I feel that this does not deliver learning.
Psychological Fidelity: But the purpose of these business simulation is learning - specifically “the extent to which the training environment prompts the essential underlying psychological processes relevant to key performance in the real-world setting” . Here the business simulation is designed to develop core business competencies - competencies such as critical thinking, business sense, strategic focus, etc. In other words the extent to which the simulation elicits the appropriate cognitive processing and through this develops the wisdom necessary to be successful. As detailed in the section entitled Beyond Reality, to create a business simulation that delivers psychological fidelity involves creating a suitably stylised and simplified model, that focuses on defined learning objectives, is suitably structured and takes into account learning process and its management.
Overall I see designing to ensure Psychological Fidelity a top down approach and designing to ensure External Validity a bottom up approach - an approach that ensures good learning.
Reality Problems impacting learning quality
Beyond Reality - the solution
I moved into simulation design for learning from designing simulation models for operations research and initially I built business simulations where I tried to replicate business behaviour (using Industrial Dynamics  concepts). But as I had to actually use these simulations for business training in the classroom I realised that modeling to achieve external validity had the problems described here.
The real world is complex with decisions interacting and outcomes influenced by internal and external factors in the short and long-term. For example,sales demand is a complex combination of price (sensitivity, knowledge, etc.), client needs, (perceived, emotional, tangible, political, etc.), product (features, availability, range, etc.), promotion (advertising, direct selling, branding, etc.), the economy, competitive actions etc. etc. etc. Learning from experience in the real word involves unraveling this complexity and understanding the links between decisions and outcomes (good and bad). Merely replicating reality retains this problem and the design of a business simulation must simplify complexity appropriately. Simplification that must be sufficient to allow learners to determine the links between their decisions and outcomes after deep thought.
The time taken to run a simulation on a course is highly correlated with the complexity of the simulation  and if sufficient time is not provided learning does not take place. A second aspect is that the time taken to develop a business simulation is highly correlated with the complexity of the simulation . When I was doing my Churchill Fellowship study I had the opportunity to sit in on the run of a simulation. It was an interesting experience because repeatedly the people who had designed the simulation had to extend the time before the decision was made because they had misjudged the time required. At each extension, the learners became more annoyed. That is not to say that in the early days (1970's) I did not make the same mistake - I did and I learned from it as I used the business simulation in the classroom..
A second aspect of complexity and the real world is that it is often confusing and unless the learners can unravel the link between the decisions they are making and the results generated they do not learn or learn the wrong thing. For example, my SMITE (Sales Management simulation) explores the management and impact of selling where the success of each individual sales person is impacted by the sales person (selling skills, client knowledge, product knowledge, morale, workload and personality), client (needs, relationship, size), competition (number, skills etc.) - making it difficult for the sales manager to take rational action. I had to work hard designing this business simulation to reduce confusion and I see a key part of this is providing information to the tutor to help him or her answer questions and manage learning.
As both simulation design time and simulation duration are highly correlated with simulation complexity replicating the real world is costly and inefficient. In other words a complex real world business simulation may last days while the learners and the sponsor wish to budget a few hours or a day. Long complex business simulations are fine for university students but not for use in the corporate classroom.
In the real world things often go wrong and this is upsetting and stressful but expected. Even if such bad experiences lead to learning in the classroom the learners do not expect this and are disengaged and unhappy learners - not a desired outcome. For example, several years ago I was discussing using business simulations and they mentioned that recently, during a simulation, they moved learners between competing teams - because "this happened in the real world" . Wondering the impact of this on engagement, I asked "will you ever do this again?" The response was no (because feedback at the end of the course was very negative). Equally, if the simulation is too complex or irrelevant then the learners will not learn and unhappy!
The following summarises key elements of my "Beyond Reality" design approach .
Learning Needs rather than replicating reality must be the starting point to designing and choosing a business simulation.
With a degree in electronics and based on my forecasting systems work with GE and extensive experience actually using business simulations to deliver learning, I see business simulation use as a dynamic process consisting of three dynamics - cognition (learning), affection (engagement) and workload . A beyond reality simulation must be designed to taking into account this dynamic process. This is particularly important for adult learners where the learning process is crucial to good, engaging learning .
Design for dynamic process involves designing and stylising an evolving learning journey.
For more information about my Systems Dynamics model
A crucial difference between the professionally trained artist and the amateur is their knowledge and use of structural composition. For example a trained artist will know about perspective, the golden section, anatomy etc. Likewise, for business simulations there are the artistic aspects - the structural composition  - how the business simulation evolves over time (Temporal-Topical System), the linkages (relationships) between decisions and results (interactions) .
The Temporal-Topical System defines how the challenges, issues, topics and environment develop as simulated time passes. Thus, it is a stylisation of the real world to ensure and provide appropriate learning. Where the Temporal-Topical System is the artistic embodiment of systems dynamics principles as the designer manages learning, engagement and work load.
The linkages between decisions and results are the second, crucial aspect of structural composition. For learning to occur, the learners must be able to tease out and understand the link between their actions (decisions) and outcomes (results). Consequentially, designing these linkages must be purposeful and besides the structural aspects there are issues of ambiguity (see later), calibration (see later) and dynamic stability (see later).
Structural Composition involves a stylisation and, often, simplification of the real world as is the case with fine art.
Outcomes in the real world are influenced by chance and this can conceal or negate the impact of decisions meaning that success (and failure) is not caused by the business person's actions. For learning to occur the learners must understand how their actions (decisions) lead to successful outcomes (results) and so the impact of chance must be managed. Equally, to ensure engagement, the learners must know that their success (or failure) is due to their actions or the actions of their competitors. If outcomes are caused by chance (random events outside the control of the learners) then the learners might not learn and become disengaged. I remember well, when working in Inventory Management with GE we had three products that typically sold 250 to 500 units a year. A salesman managed to obtain a penalty order for 33,000 units for each, to delivered in a few weeks. Happily, the production system was robust enough to do this and, I did not get my hands on the salesman (to shed blood)!
Deterministic Models - although the real world has chance events, unless these are necessary for learning randomness can and should not be incorporated. Obviously, this reduces uncertainty and ambiguity but is appropriate for many perhaps most business simulations. For example, all of my total enterprise simulations that involve direct competition between teams of learners use deterministic models. Obviously this is a stylisation of the real world but is appropriate as the simulation's purpose is to learn about business acumen or tactical or strategic leadership and not the operational management of crisis.
Stochastic Models - however, there are learning situations where chance is an inherent characteristic of the situation being modeled and must be managed by the learners. Examples of these are operations management or project management simulations. Here, the learning objective involve separating the impact of random events from desired outcomes and where the actions must reduce or eliminate the impact of random events. Here, instead of replicating the real world's random variation, it is likely to be necessary to reduce the variation, increase the impact of decisions, pre-plan random events and warn of future variation.
Consequentially, I rarely design simulations that involve chance (stochastic) elements - most of my business simulations are deterministic.
Besides designing the models and deciding decisions and results it is necessary to "calibrate" the dynamic behaviour of the business being simulated. For example when customising the Distribution Challenge simulation for Schneider Electric, the client wanted the simulation to replicate industry performance (North America Electrical Distributor Performance Analysis Report (PAR)). However, a characteristic of that industry is low profit margins and unless the learners could do better than this they would become unhappy. So, although the businesses that the learners' would run started at industry levels, the simulation was calibrated so their actions could beat the industry norms substantially. A second aspect of calibration is to ensure that (with difficulty) learners can see the linkages between decisions and results.
The simulation must be calibrated to amplify the linkages between decisions and results appropriately and make it possible for all the learners to be successful - a design stylisation.
A second aspect is where the delay causes learners to over react and the simulation to become unstable and unmanageable (literally oscillating between states). For instance my CISCO simulation (named after the Cisco Kid) replicates companies that bid for business and if they win it have to supply it. Thus it replicates industries like construction or engineering. In the real world this can lead to swings between too much work and too little work - a situation replicated perfectly by the simulation. Unfortunately this confounds learning and causes disaffection. With a lot of work by the tutors (plural) while running the simulation I overcame the problem.
Decision and result ambiguity in business simulations is a crucial design aspect. For it is ambiguity that defines the amount and depth of cognitive processing as the learners try to tease out the link between their actions (decisions) and the outcomes (results). The problem with ambiguity is that it can make understanding the links between decisions and outcomes too difficult or too easy. If too difficult learning will not take place and if too simple learners will not think deeply enough - both lead to ineffective learning, Besides ineffective learning, inappropriate ambiguity is disengaging and wastes learner time.
Decisions and outcomes in the real world have a predefined level of ambiguity. That is to say that depending on the real world situation an action or its impact may range from unambiguous to totally ambiguous. (As shown below, for the crucial decisions ambiguity in the real world tends to be very ambiguous).
So, a key design requirement is settling the appropriate degree of ambiguity based on the importance of the decision and outcome and the amount of time the learners spend on it  - a design stylisation.
You are probably asking "what is this". It relates to the detail or precision of the decision or result. For example my Executive Challenge simulation involves deciding the power of the products (and hence allows them to be positioned across a product domain) whereas the product decision for my Global Operations simulation just involves when the next of a range of products is introduced. Likewise, profitability might be shown as a percentage to two decimal places or as a comment when it is too low.
As decision and result detail impacts the amount of time required to decide or analyse (a result) settling granularity is a rational balance between the importance of the decision or result and the amount of time the learners' will spend on it - a design stylisation.
Designing reflection and conceptualisation time is crucial to good learning and means that sufficient time must be allocated to each decision-making cycle.
Having used business simulation on training courses, I know that even for a well designed and appropriate business simulation ultimately learning is ensured by the actions of the person using the business simulation (tutor). Consequentially, the design must provide information to enable the tutor to manage learning by coaching and challenging the learners. For the learners the links between decisions and results (actions and outcomes) must be appropriately ambiguous but, in contrast, for the tutor these must be unambiguous and made explicit.
Based on my experience using business simulations to deliver learning I feel that a tutor managing the process is vital.
Find out more about managing learning
Find out more about Tutor Support
Running a simulated business in a competitive environment can be highly motivating as the learners become emotionally involved with their simulated company. (For example because on residential courses the learners routinely work into the night, I never let them know my room number as otherwise I am liable to be woken in the middle of the night with questions!) But using a badly designed business simulation can be deeply disaffecting where learners feel that the learning is irrelevant, confusing etc. And, although, challenge is important, the design must ensure that this is appropriate throughout the business simulation.
This means that the challenge must progressively increase as the simulation progresses and as the learners become able to handle it.
Overall I feel that the holy grail of attempting to replicate the real world is in fact a search for fool's gold and the real 24 carat gold standard is beyond reality!
 Hall, Jeremy J. S. B. (1975) Forecasting what your business system will do, SAM Advanced Management Journal Vol. 40 No. 3 Society for Advancement of Management, New York
 Kozlowski, Steve W.J.and DeShon, Richard P. (2004) A Psychological Fidelity Approach to Simulation-Based Training: Theory, Research and Principles in SCALED WORLDS: Development, Validation and Applications Ashgate Publishing Company, Aldershot.
 Forrester, J.W. (1961) Industrial Dynamics, MIT Press
 Hall, Jeremy J. S. B. and Benita M Cox (1994) Complexity is it really that simple, Systems Developments in Business Simulations and Experiential Exercises Volume 21 eds.Precha Thavikulwat & John D. Overby, College of Business Administration, Oklahoma State University, Oklahoma
 Hall, Jeremy J. S. B. (2007) Computer Business Simulation Design: Novelty and Complexity Issues Developments in Business Simulation and Experiential Learning, Volume 34, 2007 Reprinted in the Bernie Keys Library, 8th Edition [Available from http://absel.org]
 Hall, Jeremy and Benita Cox (1993) Computerized management games:the feedback process and servomechanism analogy The Simulation and Gaming Yearbook 1993 Kogan Page London
 Knowles, Malcolm S et al (1998) The Adult Learner, Butterworth-Heinemann ISBN 0-88415-115-8
 Hall, Jeremy J. S. B. (2008) Corporate Cartooning: The Art of Computerized Business Simulation Design Developments in Business Simulation & Experiential Exercises, Volume 35, 2009, Reprinted in the Bernie Keys Library, 9th Edition [Available from http://absel.org]
 Kolb, David A. (1984) Experiential Learning: Experience as a source of learning and development Prentice Hall, Englewood Hills, NJ
 Hall, Jeremy J. S. B. (1994) Computer Paced Project Management Simulation Developments in Business Simulations and Experiential Exercises Volume 21 eds. Precha Thavikulwat & John D. Overby, College of Business Administration, Oklahoma State University, Oklahoma
 Hall, Jeremy J. S. B. (2015) Business Simulations: Reality and Beyond Developments in Business Simulation & Experiential Exercises, Volume 42, 2015,
 Hall, Jeremy J.. S. B. (2010) Beyond Reality: Aspects of Business Simulation Design and Use that Deliver Learning, Training Conference presentation San Diego California
Most recent update: 17/02/15
Hall Marketing, Studio 11, Colman's Wharf, 45 Morris Road, London E14 6PA, ENGLAND
Phone +44 (0)20 7537 2982 E-mail firstname.lastname@example.org