In an increasingly demanding environment, players in FMCG (Fast-Moving Consumer Goods) industries strive to readily provide a vast assortment of products to their clients. However, this pressures companies to keep large amounts of stock and to have an efficient production process to keep up with demand.

Our client struggled to devise a production plan that achieves the perfect tradeoff between production (e.g., utilities, human resources) and logistics (e.g., warehousing, stock holding) costs for each one of its +350 products, while considering all intricacies related to its production process.


Using a state-of-the-art mathematical optimization model, we developed a methodology capable of suggesting a 12-week production plan, which minimizes overall costs, with the necessary flexibility to accommodate the company’s dynamic production process. By considering production costs such as workforce, machine functioning, and raw-material spoilage as well as service-level and obsolescence costs, our client can now perform a data-driven decision for several of its factories.

To further streamline the company’s inventory management, we’ve also redefined the optimal safety-stock levels and production strategy, in accordance with each product’s demand patterns and required service levels.


By applying the developed methodology, our client expects to reduce total production and inventory-related costs by 4%, while experiencing a reduction of 2,5 days (9,5%)  of stock coverage due to safety stock overall reduction. 

Furthermore, by significantly reducing the time needed to develop a production plan, the company’s planning team is now able to rapidly respond to demand fluctuations and can allocate its time to more value-adding activities in the planning process.

Related Case Studies



While supply chain network design has been around for some time, organizations are now starting to acknowledge that the supply chain structure plays a too important role to be left over to chance, being able to differentiate them from the competition. Moreover, supply chains have become more complex and global due to the growing business environment volatility, as well as by virtue of the set of sophisticated technologies that have been reaching the market (and which allow the gathering of data from many different sources). So how can companies deal with these challenges and design a profitable network? A supply chain network is defined by all intervenients between the products’ origin and the final customer. The facilities’ location and their flows are of utmost importance, accounting for the majority of the supply chain associated costs and defining the service level provided to the customers. The main goal of network design is to balance these two levers and align them towards the companies’ strategic positioning, supported by robust analytical modelling. On another perspective, the complexity of the supply chain is directly linked to both internal and external factors that coexist around the business environment. The supply chain should be able not only to adjust to externalities such as changes in demand patterns or government regulations but also to accommodate the companies’ growth plans or new revenue streams (new products and/or new markets). In this insight, we explore two practical applications of network design that tackle distinct viewpoints on the topic: customer-orientation and efficiency-orientation. An efficiency-orientated approach on network design for a consumer electronics retailer Our client, a consumer electronics retailer, was facing the need of reviewing its supply chain to reduce the distribution costs. The challenge was tackled using a two-step optimization approach. Firstly, given the intended number of cross-docking hubs, the best locations to set up the operation were found. Secondly, for each configuration, operational costs were calculated using a state-of-the-art routing algorithm and the optimal number of hubs was determined. A simulator was then developed to assess the feasibility of the solutions since there were other operational details to be considered, as the variability in travelling or installation times. This validation was essential to ensure the intended service level. The optimization of the distribution network allowed the retailer to reduce its operational costs by 3%. Tackling network design in an online fashion retailer from a customer-centric standpoint The online fashion retailer aimed to assess all the impacts of opening a new warehouse with a focus on the client experience (i.e., delivery lead time, predictability) while not jeopardizing the operational distribution efficiency (i.e., shipping fees, warehousing costs). The first step of the methodology consisted on the development of a customized predictive model with the goal of identifying how the experience may impact client retention. Then a supply chain analytical model was applied to forecast shipping fees, lead times and anticipate warehousing costs. Several scenarios were outlined and the best location for the new warehouse was found. This study allowed to conclude that the opening of the new unit would increase the margin by 8%. Leverage the future performance of your supply chain As we can learn from the previous examples, irrespectively of the initial motivation behind the supply chain optimization, any network design problem needs to balance the client experience or service level against the operational efficiency. From a methodological perspective, an effective network design requires: a holistic perspective of the supply chain, with the identification of the costs, capacity levels and lead times in each point of the network; a forecasting approach, capable of foreseeing changes in the volume or customers’ expectations during a period of several years; a solid optimization or simulation approach, to evaluate different network design possibilities; a detailed roadmap towards the sustainable implementation of the new configuration.


Supply chain optimization is a topic with increasing importance, not only because efficiency is key in a progressively more competitive environment, but also because consumers are continuously more demanding about assortment diversity, lead time and price. Using analytics for supply chain optimization might just be the key to get the edge over the competition.   From understanding to decision-making To run successful supply chain analytics projects, five guidelines should be followed.   1. Understanding the operation and mapping indicators before using analytics Despite seeming an effortless task, disregarding the significance of a decision on adjacent areas is excessively common, mainly because managers are focused on their own operational scope and tend to neglect what surrounds it. Besides undermining the project’s expected results, ignoring side effects often becomes a pain point for change management in later phases.   2. Modelling the as-is operation and fully grasping its intrinsic characteristics The initial analysis frequently aims at modelling the as-is situation and comparing it with real costs and KPIs, guaranteeing that the model is a reliable representation of the reality. It is important to discuss all assumptions and outcomes with the involved stakeholders to ensure some comfort about what is and what is not being considered.   3. Mapping and exploring alternatives for further testing and evaluating While taking other players moves into consideration is obviously valuable, bringing operational and mid-level teams together to brainstorm about new possibilities is often game-changing and might result in a competitive advantage.   4. Validating the developed models and retrieving final indicators and results Considering the defined scenarios, it is time to adjust the models created to predict how each indicator would react to new supply chain configurations. Merging analytics’ prowess and business expertise is pivotal to successfully accomplish this mission. Validating the model and verifying if each lever leads to the expected result is a critical step to guarantee confidence in the results.   5. Gathering relevant stakeholders and making the final decisions Since decision-makers always have the last call, independently of the quality of the analytics project, we reinforce the importance of involving every internal and, if possible, external stakeholders. The success of the change management stage largely depends on everyone's confidence in the benefits of the project.   Tailor-made Supply Chain optimization: a case study Take the work we have recently done with a large electronics retailer, in which a challenge emerged to reshape the entire Iberian supply chain and to redesign the retailer's global network. Let us go through the five guidelines. 1. Understanding the operation and mapping indicators before using analytics In this context, the first phase of the project consisted of understanding the problem and analyzing the involved business requisites. At the same time, the team concentrated on mapping all relevant indicators (e.g., service level, lead times, etc.) and costs (rental cost, handling and stock costs for both warehouses and stores, transportation to stores and home-delivery and structure costs).   2. Modelling the as-is operation and fully grasping its intrinsic characteristics The following step of the project was to explore the available data and build an analytical model that represented the operation of the company at the time. Through statistical analysis and optimization modelling, the team was able to replicate the logistics process with close-to-reality operational indicators such as distance travelled during transportation, warehouse productivity and stock levels.   2. Mapping and exploring alternatives for further testing and evaluating Together with the retailer and according to the company’s strategic goals, a wide variety of configuration hypothesis was established for testing and evaluation. The different available solutions included having storage hubs with or without stock, keeping stock in one country or choosing a hybrid approach and outsourcing or internalizing the warehousing process.   3. Validating the developed models and retrieving final indicators and results The ensuing phase consisted in developing a holistic optimization and simulation model to test the different scenarios regarding warehouse future locations, logistic flows and warehouse-to-store allocations, aiming to minimize overall costs while ensuring a high service level. In this step, and for each scenario, all previously defined indicators and costs were predicted.   4. Gathering relevant stakeholders and making the final decisions Towards the end, all involved stakeholders were gathered, the different options were discussed and compared, and a final decision was made. The most relevant decisions of the project involved the centralization of part of the distribution process in the Portuguese warehouse, the relocation of the Spanish warehouse to a more cost-effective area, and a significant boost of the supplier-store direct deliveries. In this step, it was critical to ensure alignment between stakeholders and to reach a conclusion in which everyone agreed on, even though it wouldn’t necessarily bring gains to all parties involved. It is crucial to bear in mind that to maximize the overall benefits, some departments and stakeholders may temporarily have losses.


Most companies have now realized the importance of seeking and maintaining data, not only from their own operations but also from suppliers and customers. However, not many of them fully understand how to use that data to boost their supply chain competitiveness, by either decreasing costs or increasing customer satisfaction. As analytics becomes a popular subject, managers are broadening its range of applications within the existing corporate departments. Nevertheless, many supply chain decisions are still uniquely based on qualitative insights or competitor moves. To run successful supply chain analytics projects, there are a few guidelines that should be followed: Mapping and exploring alternatives for further testing and evaluating Understanding the operation and mapping indicators before using analytics Modelling the as-is operation and fully grasping its intrinsic characteristics Validating the developed models and retrieving final indicators and results Gathering relevant stakeholders and making the final decisions   Transforming supply chain decision-making As stated above, when performing analytical projects, it is common to make the mistake of planning modifications to organizations without considering a holistic view of its impacts. Having this thought in mind, a project was developed together with a food retailer aiming to simulate its supply chain from end-to-end, encompassing operations from the main warehouses all the way to the store shelves. The ultimate goal of this project was not to find a better supply chain configuration on itself but to give the company a tool to continuously and independently do so in the future. Therefore, the main project steps were revisited.   Mapping and exploring alternatives for further testing and evaluating The first phase of the project consisted on exploring and listing all the use cases to be covered, detailing all possible strategic decisions, tactical levers, external factors and operational variables. The defined scenarios range from high-level decisions, such as network configuration or modifying the promotional activity’s magnitude, to operational factors, such as workers’ productivity improvement.   Understanding the operation and mapping indicators before using analytics In the following step, a significant mapping effort was made to get a thorough understanding of all supply chain processes – cargo unloading, warehouse picking, store delivery, shelf replenishment, etc. –, levers – transportation delivery windows, workers and machines productivity, etc. – and variables, considering both material and informational flows.   Modelling the as-is operation and fully grasping its intrinsic characteristics The subsequent phase consisted of modelling and developing the simulator while, in parallel, a vast range of scenarios were outlined, in order to further test some of the retailer’s initiatives and action streams. This phase was particularly lengthy as model validation is a highly demanding process, implying an iterative process of both comparing model results with the reality of the operations and making further adjustments for model enhancement.   Validating the developed models and retrieving final indicators and results For each simulated scenario, several comparisons can be made with the retailer current situation, ranging from the value of store and warehouse stocks, the costs of spoilage and shrinkage to the transportation and workforce costs. The developed project and resulting tool enabled a more aggregate and holistic view of the value chain, allowing the testing of distinct scenarios and hypothesis, thus empowering a more conscious decision making and sustaining the retailer’s competitive advantage.   Gathering relevant stakeholders and making the final decisions The delivery of a tool enables the continuity of the project and leverages the developed work since it allows for the teams to carry on exploring new hypothesis in an independent manner. In this way, each team or department will be able to find, filter and decide on its own improvement opportunities, leveraging the decision-making process all through the supply chain. In summary, this project brought together business expertise and analytical knowledge, to give managers a state-of-the-art tool that allows them to automatically and continuously assess different what-if scenarios and choose the most beneficial for the company's future.


Business analytics involves several methods and tools that can be organized into three dimensions: Descriptive analytics – understanding the performance of the past (i.e., reporting) Predictive analytics – using data to anticipate how the future will look like (i.e., forecasting) Prescriptive analytics – suggesting a course of action to improve your business (i.e., optimizing)   As a manager, you should know what you want the data to do and recognize the five key benefits that business analytics yields:   1. Improved return-on-investment when compared to ‘pure analytics’ ‘Pure analytics’ means staring mining data without a specific business objective in mind. Such projects yield a high-risk of lack of results. Using business analytics, the business comes ahead of the data and guides the exploration process in a more consequent manner.   2. Superior robustness and interpretability of results when compared to ‘pure analytics’ As the business is guiding the overall process, from the problem framing to the validation of the solutions, teams are more likely to understand and use the results in the operation. Moreover, business sense should also steer away solutions that are only performing under strong assumptions.   3. More detailed decision-making when compared to just relying on business sense Without advanced analytics, managers often make decisions that work well on average, but fail to recognize the complexity of the business landscape. The power of data is connected to the possibility of tailoring decision to the different situations appropriately.   4. More alignment within your organization than with other approaches to decision making Business analytics strikes a good balance between grounding decisions on bottom-up evidence – data, while ensuring the appropriate business guidance. This equilibrium translates into team’s comfort as data brings the operational complexity with the appropriate business framing.   5. An opportunity to challenge your business beliefs With a fast-paced evolution of the expectations of the different echelons of the supply chain (e.g., suppliers and customers) it is ever more crucial to continuously challenge business beliefs that often lead to poor decision making. Business analytics, by allowing data exploration, is a very good instrument to cross-check managers intuitions.   Conclusion To reap the five key benefits that business analytics can bring, it is mandatory to have C-level support and the right team that blends business expertise and analytics prowess. The larger the number of stakeholders, who are able to manage both skills, the better.