Article

October 26, 2026

Optimization of product distribution in delivery vehicles for “e-commerce” logistics

Raphael Barretto Morgado; Igor Pinheiro de Araújo Costa

DOI: 10.22167/2675-6528-202603379

Article derived from a Course Conclusion Work (TCC), with content based on the student’s original work and adapted to the editorial format of the E&S Magazine with the support of the ResumeAI tool, an artificial intelligence solution developed by Instituto Pecege for textual synthesis and organization.

Summary

The optimization of product distribution in delivery vehicles is crucial for the competitiveness of e-commerce companies, where logistical operations represent a significant impact for improvements. The study aimed to analyze the optimization of product distribution in delivery vehicles, considering the typology of each vehicle (motorcycles, passenger cars, and utility vehicles). For this purpose, Integer Linear Programming (ILP) was used with the Pulp library from Python. Real data from an operational window and a specific cluster were transformed to create new data, preserving the original statistical relationships. Different scenarios were evaluated, varying the cost associated with each vehicle typology and the operational restriction of products per vehicle, while keeping the available fleet and the sample of packages constant. The results indicated that the distribution of vehicles per route has room for improvement, potentially reducing the number of vehicles needed to deliver all products. It was also observed that vehicle capacity was restricted by the maximum number of products allowed, although some vehicles presented volume and weight capacities below the possible limits. It was concluded that the proposed model can support strategic decisions in fleet sizing and vehicle mix definition, contributing to operational efficiency and sector competitiveness.

Keywords: Operational efficiency; Logistics; Optimization; Integer linear programming; Delivery vehicles.

1. Introduction

In the digital age, online shopping has become a routine for millions of consumers, transforming commerce and increasing business competitiveness. This scenario is corroborated by data indicating the magnitude of the phenomenon in Brazil. According to a survey by PriceWaterhouseCoopers (2021), 73% of purchases made in the country were carried out online, excluding those related to supermarkets. The Brazilian E-commerce Association (ABCOMM, 2025) points to significant growth in the total sales of these companies, which jumped from R$ 90 billion in 2019 to R$ 204 billion in 2024, representing an increase of 127%. In 2020 alone, driven by the Covid-19 pandemic, e-commerce total sales grew by 41%.

However, the growth in sales volume entails an increase in the complexity of delivery operations. Logistics represents a significant portion of companies’ costs in Brazil, corresponding to 12.4% of gross revenue, with 63.5% of this amount attributed solely to transportation, according to a study by Fundação Dom Cabral (2018). In this context, efficiency gains in transportation operations are crucial for obtaining a competitive advantage. The transportation of products in e-commerce is divided into three stages: First Mile, Middle Mile, and Last Mile. The Last Mile stage, which covers the final delivery to the customer, is considered the most challenging and costly for companies (Caetano and Gonçalves, 2023), due to factors such as the use of outsourced services, distances, hazardousness, and difficulty of access to delivery locations.

One of the main challenges in the Last Mile stage lies in defining the quantity and type of packages to be allocated to each vehicle. This complexity is aggravated by the variability in product volumes and weights, often inaccurately reported by sellers, and by the diversity of available vehicle fleets, which can include motorcycles, passenger cars, and utility vehicles like the Fiat Fiorino, each with distinct capacities. Inefficient goods allocation, resulting in underutilization of the vehicles’ total capacity, is a factor that increases operational costs and impacts the efficiency of the logistics chain.

To address optimization problems in complex scenarios like this one, Linear Programming (LP) is a classic Operations Research technique. It seeks to maximize or minimize an objective function, such as cost or profit, subject to a set of linear constraints that represent real resource limitations (Schneider, 2013). Integer Linear Programming (ILP) is an extension of LP that allows decision variables to take discrete values, being particularly useful for resource allocation problems. Previous works demonstrate the applicability of similar mathematical models in logistics network optimization, such as the location of distribution centers (Guazzelli and Cunha, 2014), blood bank location optimization (Altherwi, 2025), and freight transport optimization (Tonin, 2021).

Given the relevance of optimization in the Last Mile stage for the competitiveness of e-commerce companies, this project aimed to analyze the optimization of product distribution in delivery vehicles, considering the typology of each vehicle – motorcycles, passenger cars, and utility vehicles (Fiorino). For this purpose, Integer Linear Programming (ILP) was used with the Pulp library from Python, seeking to promote the efficient allocation of packages to available vehicles, considering restrictions on weight, volume, load capacity, and availability by vehicle category, given a determined supply of vehicles, and evaluating the impact on the utilization of transport resources and associated logistical costs, contributing to the increase in operational efficiency and sector competitiveness.

2. Material and Methods

The present study was characterized as applied research with a quantitative approach, focused on logistics optimization. The objective was to analyze the distribution of products in delivery vehicles through mathematical modeling, using Integer Linear Programming (ILP). The unit of analysis comprised product packages (weight and volume) and delivery vehicles (typology, load capacity, volume, and operational cost). The data were obtained from an e-commerce company, referring to a delivery region and operational window, simulating a specific “cluster”.

For the construction of the package database, real information on dimensions and weights was collected. To preserve the confidentiality of the original data, a linear transformation was applied, which maintained the multivariate statistical relationships but altered the scale and absolute values (Muralidhar and Sarathy, 1991). In the treatment, null values or values below 0.1 kg for weight were standardized to 0.1 kg, and for volume, a minimum of 0.001 m³ and a maximum of 4 m³ were established.

The fleet of vehicles used was synthetic, estimated from the actual availability observed on an operating day. The total number of vehicles was reduced, maintaining the proportion between types (motorcycle, passenger car, and light utility vehicle) and the relationship with the total number of packages delivered. The volume capacities of the vehicles were reduced by 10% to account for space losses. Operational costs were attributed to each vehicle category as a fixed cost, following the proportion estimated in the evaluated company.

Integer Linear Programming (ILP) was the optimization technique employed, formulated to minimize the total cost of vehicle utilization. The model sought the efficient allocation of packages within the operational limits of weight, volume, and maximum quantity per vehicle. Decision variables included x_ij (package *i* allocated to vehicle *j*) and y_j (vehicle *j* utilized). Constraints were incorporated to simulate operational conditions, where each package could be allocated at most once, and the maximum weight capacities (cap_carga[j]) and volume (cap_volume[j]) of each vehicle were considered.

Additionally, a maximum operational quantity restriction of packages per vehicle (lim_op) was applied, limiting the total number of packages that each vehicle could carry. These restrictions ensured that the allocation respected the physical and operational limitations of the fleet. Computational experiments were conducted in a laptop environment, with the model implemented in Python, using the Pandas, Numpy, and PuLP libraries.

For the resolution of Integer Linear Programming problems, the COIN-OR CBC (Computational Infrastructure for Operations Research Branch and Cut) solver was used, accessed via the PuLP library. CBC is an open-source solver that employs the Branch and Cut algorithm (Nemhauser and Wolsey, 1988). Its choice was justified by its transparency, reproducibility, and adequate performance for the objectives of this work (Mitchell, 2011; Forrest and Loungee-Heimer, 2005).

For the evaluation of the scenarios, specific indicators were defined. The average number of packages per vehicle used, the occupancy rate (ratio between occupied package volume and total available volume), and the load rate (ratio between transported weight and total weight capacity) were calculated. These indicators allowed for a standardized comparison between the scenarios.

3. Results and Discussion

The initial analysis of the real scenario, which served as the basis for comparison with the optimization model results, revealed that 845 packages were distributed using seven delivery vehicles. This operation resulted in an average of 120.7 packages allocated per vehicle. It was observed that all vehicles employed were of the “Passeio” (Passenger Car) category, with varied load and volume capacities, and that the distribution of packages among them presented different levels of utilization. For example, one passenger car carried 98.36 kg and 0.7963 m³, while another transported 178.94 kg and 0.8426 m³. This configuration, which exclusively used passenger cars, suggests that the analyzed delivery region may have a predominance of smaller-sized products, or that a fleet of vehicles with higher capacity, such as vans and utility vehicles, was not available or was not considered for this specific operation. The absence of motorcycles in the real operation is also notable, indicating a demand profile that, at that moment, did not prioritize them.

The predominance of passenger vehicles in the actual operation may indicate that the specific delivery area, or “cluster”, did not require larger vehicles for most packages, or that there was a limitation in the availability of other types of vehicles. This observation is crucial, as the inefficient allocation of goods, which results in the underutilization of the vehicles’ total capacity, is a factor that increases operational costs and impacts the efficiency of the logistics chain, as highlighted by Caetano and Gonçalves (2023). The average of 120.7 packages per vehicle, although high, may mask inefficiencies if the weight or volume capacity is not being fully utilized in all vehicles.

Optimization with Equal Costs (Scenario 1)

In the first simulated scenario, the main objective was to identify the potential for reducing the necessary fleet, removing the cost factor from the equation by assigning the same fixed cost to each vehicle modality. For this simulation, a fleet of three motorcycles, five passenger cars, and four utility vehicles was considered, totaling twelve vehicles. The operational restriction of packages per vehicle was set at 130. The results indicated that, for the same package demand, seven vehicles would still be needed, but with a fleet composition significantly different from the observed reality. The model selected one motorcycle, four passenger cars, and two utility vehicles.

The change in the fleet profile in Scenario 1 demonstrated a more efficient allocation of vehicle capacity. The selected motorcycle transported 130 packages, weighing 17.65 kg and occupying 0.1438 m³, approaching its maximum load capacity of 18 kg and volume of 0.1440 m³. Passenger vehicles, in turn, carried between 108 and 129 packages, with varied weights and volumes, but consistently below their maximum capacities of 250 kg and 1.2600 m³. The two utility vehicles (Fiorino) stood out the most in capacity utilization, transporting 119 and 120 packages, with 239.20 kg and 2.1143 m³ and 169.55 kg and 1.8698 m³, respectively, in relation to their capacity of 650 kg and 3.0105 m³. This optimized configuration, even maintaining the total number of vehicles, suggests a better adaptation of the fleet to the volume and weight of the packages, taking advantage of the diversity of available capacities.

Impact of Variable Costs (Scenario 2)

Scenario 2 introduced differentiated fixed costs for each vehicle type (1.5 for motorcycles, 2.5 for passenger cars, and 3 for utility vehicles), maintaining the operational restriction of 130 packages per vehicle. The total number of selected vehicles remained at seven, but the fleet distribution was altered compared to Scenario 1. In this scenario, the model prioritized two motorcycles, three passenger vehicles, and two utility vehicles. This change reflects the influence of costs on the allocation decision, aiming to minimize the total operation cost.

The detailed analysis of Scenario 2 revealed that the two motorcycles operated close to their maximum capacity, carrying 130 and 129 packages, with weights of 17.91 kg and 17.87 kg, and volumes of 0.1436 m³ and 0.1420 m³, respectively. The passenger vehicles showed more varied utilization, with one of them carrying only 66 packages (28.31 kg and 0.3385 m³), which is a very small load relative to its capacity. This finding suggests an opportunity for further optimization, where the load of this vehicle could be redistributed to others, or it indicates the possibility that the company is marketing higher volume products that are not properly allocated. The two utility vehicles, in turn, were well utilized, carrying 130 packages each, with weights of 322.24 kg and 246.01 kg, and volumes of 2.9863 m³ and 2.3571 m³, respectively, demonstrating efficiency in the allocation of larger packages.

Sensitivity to Operational Restriction (Scenario 3)

To evaluate the model’s sensitivity to vehicle package capacity, Scenario 3 increased the operational restriction from 130 to 150 packages, keeping costs equal for all vehicle types, similar to Scenario 1. The 15.4% increase in the limit of packages transported per vehicle generated a significant change in the required fleet, resulting in the utilization of only six vehicles, a 14.3% reduction compared to previous scenarios. The optimized fleet consisted of five passenger vehicles and one utility vehicle. This configuration demonstrates that the maximum operational package capacity is a determining factor in fleet optimization.

In this Scenario 3, passenger vehicles were allocated with a number of packages ranging between 106 and 150, with weights and volumes that approached their capacities. For example, one passenger vehicle carried 150 packages, weighing 131.47 kg and occupying 1.1998 m³, while another transported 149 packages, weighing 131.04 kg and occupying 1.1815 m³. The utility vehicle (Fiorino) was the most heavily loaded, with 148 packages, 346.16 kg, and 2.9708 m³, operating very close to its maximum volume capacity of 3.0105 m³. Most vehicles operated practically at the limit of their weight and volume capacities, with only one passenger vehicle showing some leeway. This reinforces the idea that the restriction of packages per vehicle plays a crucial role in determining the ideal fleet, allowing for greater cargo consolidation and, consequently, reducing the total number of vehicles needed to meet demand.

Combination of Costs and Constraints (Scenario 4)

Scenario 4 combined the increase in operational restriction to 150 packages per vehicle with differentiated costs per vehicle category (1.5 for motorcycles, 2.5 for passenger cars, and 3 for utility vehicles), as in Scenario 2. The variation in cost associated with the vehicle category impacted fleet selection, resulting in the use of six vehicles: two motorcycles, two passenger cars, and two utility vehicles. This configuration demonstrates that, even with increased package capacity, the cost structure still influences the choice of the most suitable vehicle types for the operation.

The motorcycles in Scenario 4 were well utilized, transporting 122 and 140 packages, with weights of 15.24 kg and 17.75 kg, and volumes of 0.1230 m³ and 0.1413 m³, respectively, operating close to their capacities. The passenger vehicles carried 149 and 135 packages, with weights of 184.43 kg and 147.75 kg, and volumes of 1.2464 m³ and 1.1524 m³, respectively, also with good utilization. The utility vehicles transported 150 and 149 packages, with 272.58 kg and 333.49 kg, and volumes of 2.8213 m³ and 2.9983 m³, respectively, demonstrating high efficiency. The model prioritized the inclusion of motorcycles and an additional utility vehicle over passenger vehicles, indicating that the cost savings generated by the use of smaller vehicles (motorcycles) outweighed the additional cost of a utility vehicle, reinforcing the perception that the profile of packages sold on the platform is predominantly small items, which can be accommodated in vehicles with lower capacity without compromising logistical efficiency.

Cost Structure Adjustment (Scenario 5)

In Scenario 5, adjustments were made to the cost weights for each vehicle type (1 for motorcycles, 2 for passenger cars, and 3 for utility vehicles), maintaining the operational restriction at 130 packages. The results showed that the selected fleet configuration was the same as in Scenario 2, i.e., seven vehicles composed of two motorcycles, three passenger cars, and two utility vehicles. This stability in fleet selection, even with the change in relative costs, suggests that, for the set of packages and observed vehicle offerings, vehicle capacity became a more determining limiting factor than the cost of each category. This indicates that, under certain conditions, the physical capacity of the vehicles is the main bottleneck, and cost optimization must operate within these capacity limitations.

The motorcycles in Scenario 5 transported 129 packages each, with weights of 17.80 kg and 17.44 kg, and volumes of 0.1415 m³ and 0.1419 m³, respectively, using almost their full capacity. The passenger vehicles carried between 67 and 130 packages, with one of them showing a significantly low load (39.39 kg and 0.3363 m³ for 67 packages), similar to what was observed in Scenario 2. The utility vehicles were well utilized, with 130 packages each, transporting 286.78 kg and 237.27 kg, and volumes of 2.9642 m³ and 2.4296 m³, respectively. The persistence of a passenger vehicle with a low load in scenarios with varied costs and constant operational restriction reinforces the need for a more in-depth analysis of the typology of products sold and the possibility of redistributing packages to maximize the utilization of all vehicles, as suggested by the literature on resource optimization (Hillier and Lieberman, 2013).

Comparative Analysis of Performance Indicators

The comparison of performance indicators between the simulated scenarios and the real scenario revealed important insights about the impacts of changes in operational assumptions. In relation to the real scenario, the number of packages per vehicle remained the same in Scenarios 1, 2, and 5, while in Scenarios 3 and 4 there was an increase of 16.7%. The load factor showed significant reductions in all optimized scenarios, ranging from -0.08% (Scenario 3) to -24.5% (Scenario 1). Similarly, the occupancy rate also decreased, with variations from -0.1% (Scenario 4) to -21.3% (Scenario 1). In the real scenario, the average vehicle occupancy rate was 96.2%.

These results indicate that optimizing fleet composition does not necessarily imply a simultaneous increase in all performance indicators. In some scenarios, fleet reorganization allowed for maintaining the same average number of packages per vehicle, but with a lower load and occupancy rate. This occurs because the introduction of vehicles with higher nominal capacity, such as utility vehicles, expands the total available fleet capacity, which, in turn, reduces the average utilization percentages, even if the solution is more suitable from a cost or operational feasibility perspective. This perspective is managerially relevant, as it demonstrates that the ideal fleet configuration must consider a balance between cost, available capacity, and the profile of the transported packages, and not just the percentage utilization of capacity (Winston, 2004).

Additionally, a considerable difference was observed in the average vehicle occupancy rate when the fleet profile is altered. Scenarios with a lower quantity of utility vehicles and motorcycles showed a more pronounced decrease in the occupancy rate. The smallest reductions occurred in scenarios that demanded fewer vehicles in total, which naturally makes them more loaded. The analysis of the occupancy rate also highlights the need for a deeper investigation into the package profile, as small increments in the volume of transported items can be decisive in choosing between a passenger car or a utility vehicle. The low reliability of package dimension data, often reported as 1x1x1 cm, leads to incorrect allocations that require manual reallocation during sorting, generating rework and inefficiency.

Managerial Implications and Model Limitations

The developed models, although they present simplifications regarding the complexity of the real problem, offer valuable managerial insights. One of the notable simplifications is the non-consideration of package geolocation in route definition, which, in theory, could lead to the allocation of packages with the same destination to different vehicles. However, this limitation is minimized by the fact that the original data sample comes from a single “cluster”, that is, a partition of the area served by a specific logistics center, reducing the probability of overlapping or inefficient routes on a large scale.

Another important simplification lies in disregarding the impact of increasing packages per delivery on the total delivery time. With more packages allocated to a single vehicle, drivers’ routes tend to extend, which can increase travel time and, consequently, customer delivery time. The delivery time variable is of great importance in the logistical context, and its inclusion in future models, along with package routing and the definition of a maximum number of stops per route, would be fundamental for a more faithful representation of operational reality (Caetano and Gonçalves, 2023).

An inherent factor to the problem’s nature, but not explicitly explored by the model, is the restriction of package types for motorcycles. Operationally, only small-sized packages (categories PP and P) are allowed for motorcycles, being separated before loading. Although the algorithm does not incorporate this rule directly, cost optimization, according to the established premises, tends to reflect the allocation of small packages to these vehicles, as they are more cost-efficient for light and low-volume loads. The analysis of the first scenario, which aimed solely at fleet reduction without considering costs, served as a reference for the lowest possible value, functioning as an ideal target. However, the model’s difficulty in simulating scenarios with many packages and vehicles limited the possibility of a real fleet reduction in a smaller cluster.

The observation that the real scenario used only passenger vehicles suggests that other delivery regions, served by the same distribution center, may require larger vehicles. This unequal distribution in package profiles between regions can impact the supply of vehicles of a certain category. From a managerial point of view, the information obtained in this work allows for strategic decisions regarding fleet sizing and vehicle distribution. Identifying the best distribution to serve a sample of packages allows for the optimal use of each type of vehicle, avoiding underutilization or overutilization. This contributes to the reduction of operational costs and minimizes the risk of undelivered routes, which generate rework at the hub.

In summary, the proposed Integer Linear Programming model demonstrated the capacity to optimize product distribution in delivery vehicles, considering different vehicle typologies and operational constraints. The results showed that the distribution of vehicles per route has room for improvement, potentially reducing the number of vehicles needed to deliver all products and optimizing fleet composition. The scenario analysis revealed that the maximum product capacity per vehicle is a crucial limiting factor, and that considering costs and package profiles is essential for strategic decisions regarding fleet sizing and vehicle mix, contributing to operational efficiency and the competitiveness of the e-commerce sector.

4. Conclusion

The present study aimed to analyze the optimization of product distribution in delivery vehicles, considering the typology of each vehicle – motorcycles, passenger cars, and utility vehicles – through Integer Linear Programming (ILP). It was found that the proposed model was capable of identifying alternative fleet configurations for “Last Mile” operations, demonstrating that vehicle distribution per route has room for improvement. It was observed that the restriction on the quantity of packages per vehicle played a relevant role in the solution, in some cases being more decisive than the weight or volume transported, allowing for a reduction in the number of vehicles needed to deliver all products in specific scenarios. The analysis of differentiated costs per vehicle category and the predominantly small profile of the packages also influenced the fleet composition, leading to more adequate solutions from the perspective of cost and operational viability. The main contribution of this work lies in supporting strategic decisions for fleet sizing and the definition of the vehicle mix, promoting operational efficiency and competitiveness in the e-commerce sector by optimizing the use of transport resources and minimizing the risks of undelivered routes.

Despite the promising results, the model presented some simplifications. The geolocation of packages in route definition was not considered, which could, in theory, lead to the allocation of packages with the same destination to different vehicles, although the sample of a single “cluster” minimized this limitation. Additionally, the impact of increasing packages per delivery on the total delivery time was not explored, nor was the operational restriction of package types for motorcycles. The low reliability of package dimension data also posed a challenge. For future studies, it is recommended to incorporate geolocation, routing, and travel time variables, as well as a maximum number of stops per route. It is also suggested to explore intermediate values for the operational restriction and a more in-depth analysis of product typology to optimize package redistribution, as well as the variation of costs per vehicle type to adapt the model to the regional fleet profile.

Bibliographic References

Associação Brasileira de Comércio Eletrônico [ABCOMM]. 2025. Principais indicadores do e-commerce. Disponível em: <https://dados.abcomm.org/crescimento-do-ecommerce-brasileiro>. Acesso em: 06 out. 2025.

PricewaterhouseCoopers Brasil. 2021. Global Consumer Insights Pulse Survey. Disponível em: < https://www.pwc.com.br/pt/estudos/setores-atividades/varejo/2021/CGIS_21_VF_pulse2.pdf>. Acesso em: 26 jan. 2026.

Resende, P. T.V.; Sousa, P. R.; Cesar, R. V.; Quintão, A. 2018. Custos logísticos no Brasil. Fundação Dom Cabral. Disponível em: <https://ci.fdc.org.br/AcervoDigital/Relat%C3%B3rios%20de%20Pesquisa/Relat%C3%B3rios%20de%20pesquisa%202018/Apresentacao_Custos_Logisticos_no%20Brasil%202018_FDC%20_%20revRVC%20abr18%20(002).pdf>. Acesso em: 13 de outubro de 2025

Article originating from the Final Course Project of the Specialization in Data Science and Analytics of the MBA USP/Esalq

To learn more about the course, click here and access the MBX Academy platform

You may also like

November 17, 2026

Detection and Diagnosis of Diseases in Corn and Rice in Mozambique with Deep Learning

Mozambican agriculture depends on corn and rice, fundamental crops for food security, but their productivity is strongly affected by phytosanitary diseases, whose early and accurate identification remains a challenge in resource-limited contexts. The study aimed to develop and compare Deep Learning-based models for the automatic detection and diagnosis of diseases in corn and rice leaves in Mozambique. Three approaches were used: a classic Convolutional Neural Network (CNN) developed from scratch, and two transfer learning models, ResNet50 and MobileNetV2. The dataset consisted of images from public databases and collected in the field, which underwent cleaning, balancing, and artificial augmentation steps. The models were trained and evaluated with standard performance metrics, including accuracy, precision, recall, F1-score, Log Loss, and Cohen’s Kappa. The classic CNN presented the best overall performance, with 94% precision, recall, and F1-score, and 93% accuracy and Cohen’s Kappa. MobileNetV2 achieved 91% F1-score and 90% accuracy, standing out for its lower Log Loss (0.30) and training time (6h32min59s). ResNet50 reached 80% F1-score and 79% accuracy. It was concluded that Deep Learning techniques have high potential to support phytopathological diagnosis in real agricultural contexts, contributing to faster and more accurate decisions in the management of corn and rice diseases.

Keywords: Deep learning; Rice; Plant diseases; Corn; Computer vision.

November 17, 2026

Artificial intelligence in accounting operational routines and financial efficiency

The incorporation of Artificial Intelligence (AI) and automation tools has reconfigured the routines of Brazilian accounting offices, impacting productivity, error mitigation, and financial efficiency. These impacts were analyzed in a small accounting firm, specialized in serving service providers in the medical field in São Caetano do Sul, São Paulo. The research adopted a qualitative approach, with a single case study design and incorporated units of analysis. Data were collected through a structured questionnaire, applied via Google Forms between January and February 2026, to five professionals in the operational structure of the firm in a census survey. The treatment followed content analysis with a priori and emergent categories. All respondents declared the use of automated tools or tools with AI resources. The perception of reduced execution time, decreased operational errors, rework, and better use of the team’s time converged in 80% of affirmative responses and 20% partial responses. The most benefited activities were bank reconciliation, import of fiscal documents, and tax assessment, with visible gains on typing errors, duplicate entries, and incorrect classification. Human intervention remained indispensable in tax planning, scenario analysis, and verification of ancillary obligations. The overall evaluation of AI’s contribution was predominantly favorable, with moderation linked to the complexity of fiscal regulations.

Keywords: Content analysis; Process automation; Accounting 4.0; Small business; Digital transformation.

November 17, 2026

Factors influencing schedule management in sewage treatment plant “retrofit” projects

Projects involving “retrofit” in Sewage Treatment Plants (STP) are characterized by high complexity, operational restrictions, and multiple technical interfaces. The study analyzed the perception of the professionals involved regarding the management and adherence to deadlines in “retrofit” projects for sewage treatment plants, considering technical, operational, and organizational factors. The research was characterized as an applied study, with a quali-quantitative approach, designed as a field survey. Data analysis was performed using descriptive statistics and qualitative categorization, employing SWOT, risk matrix (Probability x Impact), and 5W2H action plan tools. The results indicated a high perception of clarity (77.8%) and realism of deadlines (100% agreement), evidencing consistent initial planning. However, 77.7% of respondents pointed out that rework and scope changes directly impacted deadline adherence. The integration between engineering, manufacturing, and Production Planning and Control (PPC) showed an intermediate evaluation (77.7% partial/total agreement), indicating fragility in interfunctional coordination. The risk matrix highlighted as critical: scope changes, operational limitations of the STP, and technical complexity, all with a high level. As a central improvement, the standardization of scope change control, analysis, and approval was proposed, aiming to reduce rework and increase schedule predictability.

Keywords: Performance; Integration; Planning; Rework; Risk.

November 17, 2026

B2C Acquisition in diagnostic medicine: strategic partnerships with specialized medical offices

The diagnostic medicine sector in Brazil operates with high dependence on health plan providers, which limits the financial autonomy of service providers. This study proposed and analyzed a direct patient acquisition model (B2C) for a highly complex diagnostic medicine institution, structured through partnerships with Oncology and Cardiology practices and operationalized by a personalized customer service called concierge. The investigation adopted a mixed-methods approach in three phases: a field survey with a structured questionnaire applied to 87 patients, mapping of the operational flow using “Service Blueprint”, and analysis of the conversion rate from 312 medical referrals registered in the pilot project. The results indicated that trust in medical referral achieved 92.0% agreement, establishing itself as the main factor for adherence. The convenience of performing the exam on the same day reached 78.2%, and the willingness to pay directly was 55.2%, with polarization associated with the predominance of health plan users. The “Service Blueprint” identified three critical points in the transitions between stages of the patient journey. The overall conversion rate was 51.0%, with superior performance in Oncology (58.2%) compared to Cardiology (45.0%), and the greatest loss was concentrated in the acceptance and scheduling stage (28.3%). It was concluded that the model demonstrated possible operational viability and indications of patient acceptance, conditioned by price transparency and a reduction in the interval between consultation and approach by the concierge.

Keywords: Direct payment; Patient journey; Supplementary health; Concierge service; Conversion rate.

November 17, 2026

The role of the school manager in promoting an inclusive culture: leadership and educational policies

The research analyzed the school manager’s performance in promoting an inclusive culture, considering the need to transform legal guidelines and pedagogical foundations into effective institutional practices. It was assumed that school inclusion requires planning, organization, and strategic leadership in the school’s daily operations. A qualitative approach of an applied nature was adopted, developed in three complementary stages: documentary analysis of Brazilian legislation on inclusive education, bibliographic review on school leadership and management, and data systematization using management tools such as SWOT analysis and the PDCA cycle. The results indicated that the school management’s performance directly influenced the consolidation of inclusive environments, especially in the organization of processes, continuous team training, and strengthening of a collaborative institutional culture respectful of diversity. Structural, pedagogical, and relational challenges that demanded planned interventions were also identified. Finally, a Practical Guide for School Managers was developed, with guidelines, protocols, and operational orientations for decision-making and the organization of inclusive processes. It was concluded that the construction of an inclusive culture depended on the management’s ability to articulate public policies, pedagogical practices, and institutional processes, consolidating inclusion as a permanent commitment in the school context.

Keywords: educational accessibility; educational equity; participatory management; strategic planning; institutional processes.

November 17, 2026

An AI artifact based on Design Thinking to support the elicitation of software requirements from user needs

The development of software oriented to user needs frequently faces challenges in requirements elicitation, stemming from ambiguities in natural language and communication failures. Given this, an intelligent artifact integrating Design Thinking and Artificial Intelligence was proposed and validated, focusing on the Empathy and Definition stages, to support the collection, interpretation, and structuring of user needs. The research, of an applied, qualitative, and exploratory nature, conducted a Proof of Concept in a controlled environment, employing hypothetical scenarios. The artifact operated through guided natural language interviews, organizing information into elements such as pain points, context, and impact, and generating traceable problem syntheses. The results indicated that the system organized unstructured reports, reduced ambiguities, and preserved the meaning of the original information in most cases. The ability to reorganize logically in complex scenarios was also observed, although with limitations in separating multiple problems and dependence on instructions. It was concluded that the artifact has potential as a support for requirements elicitation, improving the clarity and structuring of information, but requires supervision and contextual control.

Keywords: Design Thinking; Requirements Elicitation; Requirements Engineering; Artificial Intelligence; Natural Language Processing.

November 17, 2026

The Extinction of the Monophasic Tax System in the Pharmaceutical Industry in Light of the Tax Reform

The consumption Tax Reform, established by Constitutional Amendment No. 132/2023, introduced structural changes to the Brazilian tax system, with the replacement of the Social Integration Program (PIS) and the Social Security Financing Contribution (COFINS) by the Contribution on Goods and Services (CBS), in a dual Value Added Tax (VAT) model. The objective was to analyze the impacts of the extinction of the single-stage system on the pharmaceutical industry, in light of the new regulatory design and its economic and legal effects. A qualitative, exploratory research was conducted through a case study and documentary analysis of legislation and specialized doctrine. It was examined whether the end of the single-stage system would imply a relevant increase in the tax burden for the sector. The results indicated that, although the formal overcoming of the single-stage system has occurred, the material exemption of medicines has been preserved, mainly due to the application of reduced rates of 60% and, in certain cases, a zero rate of the Tax on Goods and Services (IBS) and CBS. The change in model may cause occasional variations in revenue collection, without characterizing a structural increase in the tax burden, tending towards neutrality over time. It was preliminarily concluded that the Tax Reform reorganized the taxation technique in the pharmaceutical industry, replacing a concentrated regime with more transparent exemption mechanisms. This approach proved compatible with the principles of neutrality, full non-cumulativeness, and fiscality, maintaining tax protection for essential medicines and aligning fiscal policy with the fundamental right to health.

Keywords: CBS; COFINS; IBS; Dual VAT; PIS.