What do you do if Operations Research is leading to risky entrepreneurial decisions?

Operations Research leading to risky entrepreneurial decisions indicates that the risk factor is not included in your model, otherwise why would your optimal decisions be risky? Reformulate your model to incorporate the risk factors and then the model results would be balanced decisions. Generally high profits always come with risks, so OR model should achieve a tradeoff between these conflicting objectives.

A pesquisa operacional é como um farol para embarca??es, quando ela aponta para decis?es arriscadas, significa que o mar em que se navega até o porto é repleto de oscila??es e perigos. Em resumo a pesquisa aponta o caminho mas a forma se seguir até ele somente experiencia e prepara??o podem oferecer.

Rethink risk in your OR-driven ventures. Prioritize stress testing - evaluate solutions under various real-world scenarios to expose potential weaknesses. Integrate qualitative factors - consider social, environmental, and market risks alongside data. Foster open communication - discuss risk tolerance with entrepreneurs and adjust models accordingly. By acknowledging and mitigating risks, you can leverage Operations Research to guide sound entrepreneurial decisions.

Debemos de evaluar los reiesgos y sus impactos, para poder generar planes de reaccion y contencion en caso de un efecto negativo.

Initially implement the "risky" recommendations in parallel with the current approach. Document where the two approaches differ and follow up to see what the outcome would have been if the "risky" approach had been followed. It will become easy to evaluate the impact of the new approach as data is recorded.

Find the OR-risk equilibrium! Balance data with real-world insights. Don't let models dictate - consider human behavior and market fluctuations. Prioritize both quantitative and qualitative factors for a holistic view.

Remember the "Planner must always be in control". The "best" solution is a solution combining the combinatorial prowess of OR, with the knowledge and insights of the planner. Moreover, OR is great at crunching the numbers, but it's not so great at taking into account context that is often written or verbal in nature.

OR should always be combined with scenario comparisons (what-ifs). Scenarios can have large differences such as a completely different set of constraints provided by the planner. Or they can simply be the same constraints but with different target strategies. In the real world, all decisions are based on trade-offs and this is where OR enters a league of it's own. Balancing decisions between 2 conflicting objectives is already difficult, but balancing between 3 or more becomes near impossible for a human. OR is able to do this with the configuration of weights and priorities. So I may have multiple scenarios each with different weights and I can compare the trade-offs between each scenario

Explore beyond a single solution! Offer a range of options with varying risk profiles. Simulate scenarios with different market conditions to identify potential pitfalls. Let entrepreneurs weigh risk and reward based on their tolerance and goals. This diversification empowers them to make informed decisions using OR as a guide, not a rigid rulebook.

Explorez différentes options et scénarios possibles pour minimiser les risques. Utilisez des techniques d’optimisation pour trouver des solutions alternatives qui réduisent les risques tout en maximisant les opportunités.

Impliquez des experts et des parties prenantes clés dans le processus décisionnel. Leur perspective peut apporter un éclairage précieux sur les risques potentiels et les meilleures pratiques pour les atténuer.

Bring in external wisdom! Collaborate with industry experts familiar with the venture's domain. Their experience can identify hidden risks models might miss. Integrate their insights into your OR analysis for a more comprehensive picture. By combining data analysis with human expertise, you'll guide entrepreneurs towards well-informed and risk-mitigated decisions.

As the former owner of a consultancy, I have some bias here. However, I truly felt we brought value to every single one of our clients. Just like you might call an expert to fix an electrical or plumbing problem with your home, sometimes it's a lot less time and money to bring in an expert team to your organization rather than trying to do it yourself. If it's a capability you think your organization will need long-term, look for consultants that combine their projects with training. It was common for my company to quickly solve problems, while training staff to take ownership of the solution long-term.

Industry, but more importantly customer experience is critical. You cannot make an OR model that is supposed to support decisions if you are not able to configure the model to address the reality of the industry, customer, etc.

Organizations evolve and so the OR models must evolve with the organization to always generate the best results. This means have regular benchmarks of the performance of the models and the respect of the decisions of the model. You can do this by tracking usage of the models for starters, you may find the usage starts to drop. This could be a hint that the model is no longer being used much. But you also need to close the feedback loop. Were the models decisions followed. If not, why?

Adapt as you learn! Monitor venture progress and analyze real-world data. Identify areas where OR models need refinement. Be flexible - adjust strategies based on emerging risks and market feedback. Continuous adaptation ensures your OR approach guides sound decision-making throughout the entrepreneurial journey.

At times, risk is perceived as high, but in reality, it can be effectively managed through innovative and adaptive strategies. For instance, a significant change in policy could initially appear daunting, but with entrepreneurial intuition, opportunities for adaptation and innovation, such as offering incentives, might become apparent

Instead of radical changes, one can perhaps focus on incremental solution to the part of the problem. Stakeholder may tolerate certain risks more than the others associated with an approach to a new decision making. It will require time to see the impact of the decision and get buy in if all goes well.

There’s always difference between theoretical optimal solution and d real life operational solution. Not every solution is implementable because you didn’t include some risk factors in your model or because there are too many business rules or regulations that are not included in the model wether you failed to include them or because you omitted them so the side of the problem is manageable