登录查看更多内容

What do you do if your Operations Research analysis leads to conflicting conclusions?

由人工智能和领英社区提供技术支持

When you're deep into an Operations Research (OR) project and your analysis starts giving you mixed signals, it can feel like a major setback. But don't worry, this is a common part of the problem-solving process. Operations Research, which involves applying advanced analytical methods to help make better decisions, isn't always straightforward. If your models are spitting out conflicting conclusions, it's a sign to take a closer look. The path forward involves scrutinizing your data, models, and assumptions to understand the root of the discrepancies and how to resolve them.

此文章中的业界达人

由社区从 4 条内容中精选。了解更多

Abdelrahman S. Farrag

Graduate Research Associate | PhD Candidate in ISE
Ashish Malviya

A Digital Transformation leader with 18 Years in driving digital Innovation through AI Agent, Gen-AI, Analytics…

1 Revisit Data

If your Operations Research analysis leads to conflicting conclusions, the first step is to revisit your data. Ensure that data collection was exhaustive and free from errors. Sometimes, discrepancies arise from simple data entry mistakes or from overlooking pertinent information. Scrutinize the data sources for reliability and validity. It's also essential to verify that the data aligns with the scope of your analysis and that no critical variables have been omitted. Cleaning and validating your dataset might resolve the conflicts and lead to more consistent results.

添加您的观点

Abdelrahman S. Farrag

Graduate Research Associate | PhD Candidate in ISE
举报内容
A systematic approach is recommended. Firstly, it is crucial to revisit the system modeling and critically assess the assumptions to ensure they align with the real-world context. Secondly, one must reconsider the data inputs to confirm their validity and reliability, as noisy or incomplete data can lead to misleading outcomes. Third, conduct a sensitivity analysis and it can help identify which assumptions or inputs most influence the outcomes. Finally, incorporating realism through multiple discussions with stakeholders is essential. These discussions can provide additional insights and help refine the objectives or constraints of the model to ensure that the model is both practical and accurate.

已翻译

赞
Ashish Malviya

A Digital Transformation leader with 18 Years in driving digital Innovation through AI Agent, Gen-AI, Analytics, Industry 4.0, Digital Transformation, IT, Smart Factory, IoT, AI & BI for Business Growth and Innovation.
举报内容
If your Operations Research analysis yields conflicting conclusions, particularly in IIoT and MES contexts, reevaluate your data and models. Verify the accuracy of data inputs and consistency of parameters across different scenarios. Use sensitivity analysis to identify which variables most influence the outcomes. Consider consulting with stakeholders or experts to gain insights that might explain discrepancies. Integrating additional data sources or adjusting modeling techniques can also provide clearer insights. This systematic approach ensures robust decision-making for your Industry 4.0 initiatives.

已翻译

赞
Michael Agarana, Ph.D.

Professor at Lambton College, ON. Canada Associate Professor of Mathematics. Operations Research/Management Consultant. Swarm/Artificial Intelligence Researcher. ROBOTICS/STEM Enthusiast.
举报内容
Operations Research analysis are bound to lead to conflicting conclusions if the data is biased and the parameters and variables are not properly defined. That is, not giving the correct or exact meaning or value. When such inputs are used for analysis, the results will be misleading. So, the data have to be good and the parameters and variables in the model have to be well and correctly defined.

已翻译

赞

2 Model Check

After ensuring your data is accurate, examine the models used in your analysis. Operations Research relies on mathematical models to simulate real-world scenarios, and if these models are incorrectly specified or calibrated, they can yield conflicting results. Check for the correct application of algorithms and the appropriateness of the models for the problem at hand. It's possible that different models or approaches are suitable for different parts of your dataset or problem, so consider whether a combination of models could offer a more coherent conclusion.

添加您的观点

3 Challenge Assumptions

Every Operations Research model is built on a set of assumptions. When faced with conflicting conclusions, challenge these assumptions. Are they still valid? Could changes in the external environment have rendered some assumptions obsolete? Sometimes, what was considered a constant can change, leading to discrepancies in outcomes. By critically evaluating your assumptions, you can adjust your models to better reflect current conditions and potentially resolve the conflicting findings.

添加您的观点

Michael Schilde, PhD

??♂? Building Digital Bridges for SMEs | Certified Digital Consultant | IT & Process Efficiency & Scalability through Smart Digitalization
举报内容
Aus meiner Erfahrung heraus, stehen Menschen oft vor dem Problem, dass ihre Annahmen auf einer gewissen Erwartung oder Hoffnung basieren. Angesichts dessen f?llt es vielen sehr schwer, sich angesichts widersprüchlicher Schlussfolgerungen damit abzufinden, dass ihre ursprüngliche Annahme durch die Daten nicht best?tigt wurde. Hier ist es, wie auch bei allen anderen datengetriebenen Aufgabenstellungen, essenziell, zu akzeptieren, dass Hypothesen durch Daten auch widerlegt werden k?nnen - ganz egal, wie logisch zwingend oder "intuitiv" sie auch gewesen sein m?gen. Es hei?t nicht umsonst "datengetrieben". Wenn man das nicht beherzigt, belügt man sich nur selbst. Man muss lernen, es als das akzeptieren, was es ist: ein Erkenntnisgewinn.

已翻译

赞

4 Seek Expertise

Operations Research is complex, and sometimes you need a fresh set of eyes. Seeking expertise from colleagues or industry experts can provide new insights into why your analysis is conflicting. They may spot errors or suggest alternative approaches that you hadn't considered. Collaboration is a key aspect of OR, and leveraging diverse expertise can help in reconciling conflicting results and strengthening the overall analysis.

添加您的观点

5 Sensitivity Analysis

Performing a sensitivity analysis can help identify which inputs have the most significant impact on your model's outcomes. By systematically varying key parameters within your model, you can observe how sensitive the results are to changes. This process can highlight which variables or assumptions are causing the conflicts and guide you towards areas that may need more robust data or a different modeling approach.

添加您的观点

6 Iterative Process

Remember that Operations Research is an iterative process. If your analysis leads to conflicting conclusions, it's an opportunity to learn and improve your approach. Revisit each step of your analysis, from data collection to model selection and assumption testing. Each iteration should bring you closer to resolving the conflicts and finding a solution that is robust and reliable. Embrace the iterative nature of OR as a path to deeper understanding and better decision-making.

添加您的观点

7 Here’s what else to consider

This is a space to share examples, stories, or insights that don’t fit into any of the previous sections. What else would you like to add?

添加您的观点

Operations Research

+ 关注

给文章评分

我们借助人工智能创建了此文章。您认为这篇文章怎么样？

很棒不太好

举报此文章

查看全部

What do you do if your Operations Research analysis leads to conflicting conclusions?

1

2

3

4

5

6

7

1 Revisit Data

2 Model Check

3 Challenge Assumptions

4 Seek Expertise

5 Sensitivity Analysis

6 Iterative Process

7 Here’s what else to consider

Operations Research

给文章评分

感谢您的反馈

更多Operations Research相关文章

更多相关阅读内容

What do you do if your Operations Research analysis leads to conflicting conclusions?

1

2

3

4

5

6

7

1 Revisit Data

2 Model Check

3 Challenge Assumptions

4 Seek Expertise

5 Sensitivity Analysis

6 Iterative Process

7 Here’s what else to consider

Operations Research

给文章评分

感谢您的反馈

查看其他技能