Samenvatting
Samenvatting Behavioral Operations Management
- Instelling
- Technische Universiteit Eindhoven (TUE)
Samenvatting van de lectures + alle bijbehorende artikelen
[Meer zien]Voorbeeld 4 van de 54 pagina's
Voorbeeld 4 van de 54 pagina's
In winkelwagengesponsord bericht van onze partner
Voorbeeld van de inhoud
Behavioral Operations Management – 1JM40 – 2021/2022 – Q3Topic 1: Introduction and behavioral forecasting
Lecture notes
Introduction to Behavioral Operations Management (BOM)
What is Operations Management (OM)? a field of investigating the design, management and
improvement of processes aimed at the development, production, delivery and distribution of products
and services
- In the majority of these processes, people are a critical component of the system (e.g. sales
forecasting), influencing the system’s performance
- OM largely ignored this human factor as a result of some assumptions this field holds
regarding the determinants of human behavior (i.e. people who participate in these processes
are ‘rational agents’)
Key assumption of OM: rational agents
A rational agent is an agent who has clear preferences, models uncertainty via expected values, and
always chooses to perform the action with the optimal expected outcome from among all feasible
actions.
This view of human nature leads to the following predictions
- People are self-interested (and believe others are also self-interested)
- People have stable preferences
- People can distinguish signal from noise (they only respond to relevant information and are
able to discard irrelevant information)
- People’s decisions are unaffected by cognitive biases or emotions
For many years, OM built their models on the assumption that individuals are rational. Based on this
idealized notion of human decision-makers, OM largely ignored this ‘human factor’ when designing,
managing and improving operational processes. This lack of attention is noteworthy given OM’s
history: scientific management (Taylor) heavily focused on optimizing production processes by
reorganizing how people work.
- Although Taylor was aware of people’s limitations, OM shifted its focus and adopted the
assumptions that people are rational (and thus not a component to take into account)
- These assumptions were challenged by (among others) work of Simon and
Tversky&Kahneman, showing that people are limited in their ability to process information
people rely on heuristics and are affected by cognitive biases
Behavioral operations management (BOM) explicitly acknowledges that people are guided by
emotions, cognitive biases or irrelevant situation cues when making decisions definition: BOM is
the study of human behavior and their impact on the design, management and improvement of
processes aimed at the development, production, delivery and distribution of products and services.
people often deviate from superior models
OM & BOM share the same overall goal, btu their focus is
different: BOM focusses on how cognitive and behavioral
factors shape the way operational processes work and
perform. Why? understanding the role of human
behavior in operations can help us to design processes that
account for the behavioral tendencies of managers,
workers and customers.
How do people behave? cognitive biases and heuristics
• Biases are systematic errors that affect people’s decisions and judgments
• Heuristics are rules of thumb that people commonly employ to navigate the ocean of
information available to them
Examples
Anchoring and adjustment heuristic: people’s tendency to rely too heavily (or anchor) on one
piece of information when making decisions.
1
, o Example: Participants observed a wheel of fortune that was predetermined to stop on either 10
or 65. Participants were then asked to guess the percentage of the United Nations that were
African nations. Participants whose wheel stopped on 10 guessed lower values (25% on
average) than participants whose wheel stopped at 65 (45% on average). People evidently
used the prior number (10 or 65) as anchor when answering the subsequent question.
can also be used in sales forecasting since when you get a specific number you will
deviate from that to make the next forecast
Planning fallacy: people’s tendency to underestimate task-completion times
Example: Students were asked to estimate how long it would take to finish their thesis. The average
estimate was 33.9 days. The average actual completion time was 55.5 days. Possibly explained by
wishful thinking, overconfidence, focalism
Real life example of Berlin Brandenburg Airport : start construction 2006 (initial opening 2011)
opened in October 2020. Reasons delay: poor construction planning, execution, management, and
corruption.
Survivor bias: people’s tendency to concentrate on things that made it past some selection
process and overlooking those that did not, typically because of their lack of visibility.
Example: During World War II, the Center for Naval Analyses had conducted a study of the damage
done to aircraft that had returned from missions, and had recommended that armor be added to the
damaged areas. The statistician Abraham Wald noted that the study only considered the aircraft that
had survived their missions and proposed that the Navy should reinforce areas where the returning
aircraft were undamaged.
eg customer satisfaction typically you send a survey to current customers, also good to
send a survey to prior customers to understand why they left your store
In understanding the role of human behavior in operations, people working in BOM use a variety of
research methods to test their hypotheses common research methods
1. Experiments
2. Secondary data (e.g. company log-files)
Experiments are studies in which a researcher manipulates a variable in order to observe how it
effects another variable that is being studies
- Several experimental conditions
- Random allocation of individuals to one of the conditions
- Conditions are equal, except for the variation of the independent variables
Advantages: experiments allow you to make ‘causal claims’
Disadvantage: low degree of external validity – difficult to generalize to other situations (experiments
are often artificial – increased interest in field experiments, but difficult to accomplish + often not
possible to randomly allocate participants to different conditions)
Eg how people do respond to algorithms do people immediately respond from the algorithm or
after a slight delay? On what extend do they rely on this predication? the predictions are the same
but they trust the prediction more after the delay, but in practice this should not be the case.
Experiments are used to see if this is really true and to make ‘causal claims’ Example of
experiment is paying people to reward (article will come back later)
Secondary data is information that has already been collected and recorded by someone else
(companies in our context), usually for other purposes.
Advantage: realistic situation, high degree of external validity (results apply to circumstances beyond
those studied)
Disadvantage: impossible to make causal claims (hard to tell what caused an observed relationship,
e.g. is observation that planner more often adjust the statistical forecast upwards explained by
optimism bias or stock out aversion?)
Behavioral forecasting
Sales forecasting is the process of estimating future sales and forms a crucial aspect of the planning
process.
- Accurate forecasting results in significant financial savings, higher competitiveness, better
supply chain relationships and improved customer satisfactions (Moon, Mentzer & Smith, 2003).
- Three main approaches in forecasting future sales of products
o Pure judgmental forecasting: planners make a forecast without a decision aid
o Combining forecasts: separate system forecast + judgmental forecast are combined
2
, o Initial baseline statistical forecast + adjustment by planner to account for exceptional
circumstances
Key questions:
• Making these adjustments involve considerable effort and time, but do they improve the
accuracy, and are some types of adjustments more effective than others? Fildes et al.
(2009)
• How can we effectively combine human judgment with the solution of a system? Fildes et
al. (2009)
Example of own research: large multinational which has statistical forecasts for a large number of
products. First part: findings on the role of human decision-making in sales forecasting: focus on
uncovering behavioral regularities that are predictable and observable; the consequence (in terms of
forecasting error) of these regularities. Second part: given these regularities, how can we improve the
forecasting process? Key notion: this process is certainly possible as human decision-making is
predictable inaccurate for some tasks while highly accurate when performing other tasks.
Research questions and conclusions also see slides
1. How often do sales planners adjust a statistical forecast?
Sales planners adjust the statistical forecast on a regular basis (36% of the time the forecast
is adjusted). Somewhat lower than what we have seen in other companies – may have to do
with the number of products in a planner’s portfolio.
2. In what direction do planner adjust a statistical forecast?
Sales planner are inclined to make an upward adjustment (79% upward). Upward
adjustments are also larger (planners increase the statistical forecast, on average, by adding
approximately 90% while lowering the statistical forecasting by approximately 25% when
moving downwards). the tendency to make upward adjustment is very robust (seen in
every region)
3. Is the adjustment an improvement?
In order to determine whether the adjustment is an improvement, we need to compute the
forecasting error for which we use mean absolute error.
o Error statistical forecast = (abs) actual sales – statistical forecast
o Error forecast planner = (abs) actual sales – adjusted forecast
On a general level: as compared to the initial statistical forecast, adjusting the forecast does
not lower the error (adjustments add approx. 10% error). Conflicting findings in the literature
on whether adjustments increases or decrease error – Fildes et al show that adjustments
decrease error in 3 out of 4 companies. We typically see that adjustments increase error.
4. Does the (in)accuracy of an adjustment depend on the direction of the adjustment?
When making an upward adjustment, the adjustment produces more error (+19%). When
making a downward adjustment, the adjustment produces less error (-14%).
Very robust pattern true for every region, country
5. When adjusting the statistical forecast, is the planners adjusting it in the right direction?
Sales planners are reasonably good in adjusting the forecast in the right direction (55%).
They still make many mistakes (note that mistakes in direction by definition lower the
accuracy of the adjusted forecast as compared to the statistical forecast)
6. When are planners good in adjusting the forecast in the right direction?
Sales planners are especially accurate in predicting that the statistical forecast should be
lower. Mistakes in direction are often made when adjusting the statistical forecast upward.
Summary first part (impact human behavior)
- Sales planner quite often adjust the statistical forecast (mostly upwards)
- These adjustment generally do not lower the forecasting error
- Sales planners are especially inaccurate, as compared to the system, when adjusting
upwards (adds error often adjustments in wrong direction very stable pattern)
3
, - Sales planners outperform the system when making downward adjustments (lower error
often adjust in the right direction stable pattern)
- Regional and BU differences in overall error (although behavior patterns are very similar)
Summary second part (testing new method of
forecasting)
What if sales planners decide on the direction of
the adjustment (up vs down) while the algorithm
decides on the magnitude of the adjustment
(especially when sales planners indicates that
the statistical forecast should go up?)
In this case: when planners says up, to with the
statistical forecast +5%, when planner says
down, go with the downward adjusted forecast
see slides for results of testing algorithm
result = more efficient, promising to reduce the forecast error
Gino&Pisano (2008). Toward a theory of behavioral operations.
Introduction
The enduring importance of human behavior in operations suggest that people may significantly
influence how operating systems work, perform and respond to management interventions. Most
formal analytical models in operations management (OM) assume that the agents who participate in
operating systems or processes – as decision makers, problem solvers, implementers, workers, or
customers – are either fully rational or can be induced to behave rationally.
Intellectual terrain of OM
Om is a multidisciplinary field that investigates the design, management and improvement of
processes aimed at the development, production and delivery, and distribution of products and
services. Simon (1957): bounded rationality the capacity of the human mind for formulating and
solving complex problems is very small compared with the size of the problems whose solution is
required for objectively rational behavior in the real world – or even for a reasonable approximation to
such objective rationality.
Closer look at human behavior in other fields
- For many years, economists built their models on the assumption that individuals are rational.
- Simon and Tversky&Kahneman argued that human beings are limited in their capacities to
learn, think and act, and that these limits have important implications for economic theory.
- Biases results from cognitive limitations and are systematic errors that affect people’s
decisions or judgements
- Heuristics are rules of thumb that people commonly employ to navigate the ocean of
information available to them in their decision-making process
Behavioral operations: a definition
Behavioral operations: the study of human behavior and cognition and their impacts on operating
systems and processes.
- BOM & OM share the same goal but their research focus is different BOM treats human
behavior as a core part of the functioning and performance of operating systems
Opportunities for behavioral research in OM
Two main approaches in which behavioral considerations can be included in OM models
1. Prescriptive: behavioral factors should be integrated into OM models
2. Descriptive: highlights the relevance of understanding the effects of individuals’ shortcomings
on decision-making processes and problem-solving activities within OM contexts
Rationale for behavioral operations: should be improved insight into and understanding of problems
use of anomalies since they provide a useful starting point for the construction of a new theory
example = bullwhip effect
See Table 1 in article for different impacts of heuristics and biases on operating systems and
processes
Conclusions
- A behavioral approach to OM can lead to a better understanding of underlying drivers of
operating system performance and also to a better understanding of puzzling ‘pathologies’
4
Voordelen van het kopen van samenvattingen bij Stuvia op een rij:
Verzekerd van kwaliteit door reviews
Stuvia-klanten hebben meer dan 700.000 samenvattingen beoordeeld. Zo weet je zeker dat je de beste documenten koopt!
Snel en makkelijk kopen
Je betaalt supersnel en eenmalig met iDeal, creditcard of Stuvia-tegoed voor de samenvatting. Zonder lidmaatschap.
Focus op de essentie
Samenvattingen worden geschreven voor en door anderen. Daarom zijn de samenvattingen altijd betrouwbaar en actueel. Zo kom je snel tot de kern!
Veelgestelde vragen
Wat krijg ik als ik dit document koop?
Je krijgt een PDF, die direct beschikbaar is na je aankoop. Het gekochte document is altijd, overal en oneindig toegankelijk via je profiel.
Tevredenheidsgarantie: hoe werkt dat?
Onze tevredenheidsgarantie zorgt ervoor dat je altijd een studiedocument vindt dat goed bij je past. Je vult een formulier in en onze klantenservice regelt de rest.
Van wie koop ik deze samenvatting?
Stuvia is een marktplaats, je koop dit document dus niet van ons, maar van verkoper daniekklijn. Stuvia faciliteert de betaling aan de verkoper.
Zit ik meteen vast aan een abonnement?
Nee, je koopt alleen deze samenvatting voor €4,99. Je zit daarna nergens aan vast.
Is Stuvia te vertrouwen?
4,6 sterren op Google & Trustpilot (+1000 reviews)
Afgelopen 30 dagen zijn er 52355 samenvattingen verkocht
Opgericht in 2010, al 14 jaar dé plek om samenvattingen te kopen