I am thinking about a question of Behavioral Operations Management (BOM): how to define a contribution to the literature? For the traditional operational research, it can be a new model, a new algorithm or a model for a new problem. But what is a contribution in BOM? In BOM, we first observe the system biased performance (primary), then look into the biased decisions which lead to the performance (secondary), try to find out what biases induce such decisions and how (tertiary), and is it necessary to go deeper to explore reasons for the biases (fourth)? I found it is hard to seek for explanations in the fourth level. However, without the fourth level, how can we propose suggestions to adjust these biased behaviors in the decision process? Without efficient suggestions, what is the contribution of a research? Or, am I wrong with the definition of contribution in BOM? How far should we go in this field?
Hi Yingshau - First, let's stick to the more standard BeOps (BOM=Bill of Materials).
Second, you're thinking about this from the perspective of someone who has a model and just wants to correct its problems in application (which of course is something that has dogged OR for decades). Instead try to think of it as foundational. Why is the history of psychology and sociology research what it is? Do these fields continue to contribute to practice? Of course. Can we assume the findings from studies in these fields are robustly applicable in all OM contexts? Certainly not, since they don't tend to study OM contexts.
Hence the disconnect. OR models have a long track record of applicability failures, particularly when key elements are humans (as is almost always the case). Psych and sociology have made great strides... ones that general management, marketing and IS have strongly embraced and benefits from. This is what BeOps has done for operations. It's started us down the road towards understanding better what we would like to be able to practice with... only in a much more realistic way.
The question of "what's next" is as broad as the fields of psychology and sociology, and as ample as has been the applications in other disciplines, As researchers we just need to be a little more open to these things, and less willing to fall back on incremental "improvements" to the field. It's precisely the shot we've needed.
Return to the basics. Observation, listening, and variance to the standard must apply. Take for instance what was observed a left handed person work from the right side of a production line compared to a right handed person. The flow of work is different. Can it be done from the other side taking all things into consideration - safety, material, egress?
Language instructions apply. It would be of interest to study the difference between written work orders verses audio work orders. ISO standards call for detailed work instructions - how about computer based work instructions that are interactive. A call center for technical assistance. The use of colored lights in my opinion demeans the individual. Could the use of a difference visual tool or audio sound be used to acknowledge when a line is meeting production targets, exceeds, or is behind. Audio would provide acknowledgement to supervision that might be engaged in other activities ( a code blue).
It is important to continue to observe, listen, engage, and deviate from standard.
@Elliot Bendoly Prof. Bendoly, thank you very much for your answer. It is helpful. The main idea of the answer is that "BeOps is the foundation of extending OM in a more realistic ways. From this perspective, the future research of BeOps is broad and full of opportunities." Please let me know if I am wrong.
I thought about it for days, but still have problems with treating BeOps as a foundation. When I observe a phenomenon (e.g., orders are increased when purchasing cost is decreased, while the increasing rate decreases as the purchasing cost goes on decreasing), should I explain the bias by organizing another specially designed experiment (so far I have focused on experimental methodology), or by the principles proposed in psychology and sociology? They both seem applicable, but are quite different.
Or should I just propose the observation of the biased phenomenon (without any explanation), and suggest managers to be careful with the biases in certain context?
The above contains two questions: i) How to explain biases in OM practice; ii) Are observations/suggestions enough to be research contributions in BeOps?
Now, the theoretical benchmark came to my mind again. I cannot help comparing the BeOps with the traditional OR. Could you please give me some suggestions about changing my thinking method?
I appreciate your time and answer very much.
@William Lynch Thank you for your answer. It inspires a lot. "Observation, listening, and variance to the standard must apply." It is very important.
The next questions should be: "Is standard right or wrong?" "Is deviation from standard better or worse?"
It can be verified in some ways. To be greedy, I want to know the reason. I think I should know the reason so that I can do induction and deduction, but no limit to a specified scenario.
However, I have read an interesting history book recently. It says that events happened because of the interactions of many earlier events. I guess the behavior of decision makers is also influenced by factors in this way. Considering the interaction, it increases the complexity of explaining the deviation. I'm wondering how to solve it.
If you can provide any suggestion, I will appreciate your help very much. Thank you for reading.
@Elliot Bendoly Thank you for the information, professor. Via the website, I found another link (i.e., https://sites.google.com/site/ombehavior/home) is also very helpful. Many thanks!