Abstract
Sales forecasts are an important tool for inventory management, allowing retailers to balance inventory levels with customer demand and market conditions. By using sales forecasts to inform inventory management decisions, companies can optimize their inventory levels and avoid costly stockouts or excess inventory costs. The scale of the forecasting problem in the retail domain is significant and requires ongoing attention and resources to ensure accurate and effective forecasting. Recent advances in machine learning algorithms such as deep learning have made possible to build more sophisticated forecasting models that can learn from large amounts of data. These global models can capture complex patterns and relationships in the data and predict demand across multiple regions and product categories. In this paper, we investigate the cross-learning scenarios, inspired by the product hierarchy frequently utilized in retail planning, which enable global models to better capture interdependencies between different products and regions. Our empirical results obtained using M5 competition dataset indicate that the cross-learning approaches exhibit significantly superior performance compared to local forecasting benchmarks. Our findings also suggest that using partial pooling at the lowest aggregation level of the retail hierarchical allows for a more effective capture of the distinct characteristics of each group.

Abstract
Global models have been developed to tackle the challenge of forecasting sets of series that are related or share similarities, but they have not been developed for heterogeneous datasets. Various methods of partitioning by relatedness have been introduced to enhance the similarities of sets, resulting in improved forecasting accuracy but often at the cost of a reduced sample size, which could be harmful. To shed light on how the relatedness between series impacts the effectiveness of global models in real-world demand-forecasting problems, we perform an extensive empirical study using the M5 competition dataset. We examine cross-learning scenarios driven by the product hierarchy commonly employed in retail planning to allow global models to capture interdependencies across products and regions more effectively. Our findings show that global models outperform state-of-the-art local benchmarks by a considerable margin, indicating that they are not inherently more limited than local models and can handle unrelated time-series data effectively. The accuracy of data-partitioning approaches increases as the sizes of the data pools and the models' complexity decrease. However, there is a trade-off between data availability and data relatedness. Smaller data pools lead to increased similarity among time series, making it easier to capture cross-product and cross-region dependencies, but this comes at the cost of a reduced sample, which may not be beneficial. Finally, it is worth noting that the successful implementation of global models for heterogeneous datasets can significantly impact forecasting practice.

Abstract
Retailers depend on accurate forecasts of product sales at the Store x SKU level to efficiently manage their inventory. Consequently, there has been increasing interest in identifying more advanced statistical techniques that lead to accuracy improvements. However, the inclusion of multiple drivers affecting demand into commonly used ARIMA and ETS models is not straightforward, particularly when many explanatory variables are available. Moreover, regularization regression models that shrink the model's parameters allow for the inclusion of a lot of relevant information but do not intrinsically handle the dynamics of the demand. These problems have not been addressed by previous studies. Nevertheless, multiple simultaneous effects interacting are common in retailing. To be successful, any approach needs to be automatic, robust and efficiently scaleable. In this study, we design novel approaches to forecast retailer product sales taking into account the main drivers which affect SKU demand at store level. To address the variable selection challenge, the use of dimensionality reduction via principal components analysis (PCA) and shrinkage estimators was investigated. The empirical results, using a case study of supermarket sales in Portugal, show that both PCA and shrinkage are useful and result in gains in forecast accuracy in the order of 10% over benchmarks while offering insights on the impact of promotions. Focusing on the promotional periods, PCA-based models perform strongly, while shrinkage estimators over-shrink. For the non-promotional periods, shrinkage estimators significantly outperform the alternatives.

Abstract
Purpose - Humanity and all life depend on the natural environment of Planet Earth, and that environment is in acute crisis across land, sea and air. One of a set of commentaries on how service can address the UN's sustainable development goals (SDGs), the authors focus on environmental goals SDG 13 (climate action), SDG 14 (life below water) and SDG 15 (life on land). This paper aims to propose a conceptual framework that incorporates the natural environment into transformative services. Design/methodology/approach - The authors trace the evolution of service thinking about the natural environment, from a stewardship perspective of the environment as a set of resources to be managed, through an acknowledgement of nonhuman organisms as actors that can participate in service exchange, towards an emergent concept of ecosystems as integrating human social actors and other biological actors who engage fully in value co-creation. Findings - The authors derive a framework integrating human and other life forms as co-creating actors, drawing on shared natural resources to achieve mutualism, where each actor can have a net benefit from the relationship. Future research questions are posited that may help services research address SDGs 13-15. Originality/value - The framework integrates ideas from environmental ecosystem literature to inform the nature of ecosystems. By integrating environmental actors and ecological insights into the understanding of service ecosystems, service scholars are well placed to make unique contributions to the global challenge of creating a sustainable future.

Abstract
Background. Traditional burning is a practice with social and ecological value used worldwide. However, given the often improper and negligent use of fire, this practice is often associated with rural fire ignitions.Aims. Systematise experts' understanding of traditional burning and identify its challenges in the Portuguese context.Methods. Twenty-eight Portuguese experts from industry, academia, NGOs and public entities with in-depth involvement in fire and forest management were interviewed to create a mental model of traditional burning in Portugal.Key results. Eight dimensions were identified: motivations behind traditional burning, alternative solutions, risks before a traditional burn, risks during a traditional burn, underlying causes of risk, exogenous elements and factors, potential impacts, and activities leading to a successful traditional burn.Conclusions. This study provides a comprehensive understanding of traditional burn practice in the Portuguese context and offers a baseline to support stakeholders and policymakers in managing traditional burning's social and environmental impacts in the future.Implications. This research offers several implications across the eight dimensions identified, including the need to improve regulations on the use of fire and fuel reduction policies, promote fire use education and feasible and affordable alternatives to traditional burning, and increase communities' commitment to mitigation actions.

Abstract
PurposeWhile academics and practitioners increasingly recognize the impacts of gamification on customer experience (CX), its role in the customer journey remains undeveloped. This article aims to identify how gamification can leverage each customer journey stage, integrate the findings into a conceptual model and propose future research opportunities.Design/methodology/approachSince CX and customer journey are interrelated concepts, the authors rely on CX research to identify research themes that provide insights to propose the conceptual model. A systematic review of 154 articles on the interplay between gamification and CX research published from 2013 to 2022 was performed and analyzed by thematic content analysis. The authors interpreted the results according to the service customer journey stages and the taxonomy of digital engagement practices.FindingsThis article identified five main thematic categories that shape the conceptual model (design, customer journey stages, customer, technology and context). Gamification design can support customer value creation at any customer journey stage. While gamification can leverage brand engagement at the pre-service stage by enhancing customer motivation and information search, it can leverage service and brand engagement at the core and post-service stages by enhancing customer participation and brand relationships. Moreover, customer-, technology- and context-related factors influence the gamified service experience in the customer journey.Originality/valueThis article contributes to a conceptual integration between gamification and customer journey. Additionally, it provides opportunities for future research from a customer journey perspective.