• Data-driven vehicle routing with profits

      Vercamer, D.; Baecke, Philippe; Gendreau, M.; Van den Poel, Dirk (2015)
    • Developing a maturity assessment model for demand forecasting

      Vereecke, Ann; Vanderheyden, Karlien; Baecke, Philippe; Van Steendam, Tom (2016)
    • From one-class to two-class classification by incorporating expert knowledge

      Oosterlinck, Dieter; Benoît, Dries; Baecke, Philippe (2018)
      In certain business cases the aim is to identify observations that deviate from an identified normal behaviour. It is often the case that only instances of the normal class are known, whereas so called novelties are undiscovered. Novelty detection or anomaly detection approaches usually apply methods from the field of outlier detection. However, anomalies are not always outliers and outliers are not always anomalies. The standard one-class classification approaches therefore underperform in many real business cases. Drawing upon literature about incorporating expert knowledge,we come up with a new method that significantly improves the predictive performance of a one-class model. Combining the available data and expert knowledge about potential anomalies enables us to create synthetic novelties. The latter are incorporated into a standard two-class predictive model. Based on a telco dataset, we prove that our synthetic two-class model clearly outperforms a standard one-class model on the synthetic dataset. In a next step the model was applied to real data. Top identified novelties were manually checked by experts. The results indicate that incorporating expert knowledge to transform a one-class problem into a two-class problem is a valuable method.
    • Improving profitability of vehicle routing problems through advanced analytics

      Vercamer, Dauwe; Van den Poel, Dirk; Gendreau, Michel; Baecke, Philippe (2014)
      Based on a real case in door-to-door sales, the study assesses whether revenue predictions coming from transactional data can effectively be used to improve fleet schedules. To do this, two different customer selection models are compared. In the static model, customers are first chosen based on the revenue predictions and then routed through a VNS. The dynamic model uses the predictions in an orienteering problem. Initial results show that the dynamic approach is the most profitable.
    • Social ties in customer referral programs

      Roelens, Iris; Baecke, Philippe; Benoît, Dries (2018)
      Customer referral programs are marketing programs in which existing customers are rewarded for bringing in new customers. The aim is to attract new customers by leveraging the social connections of existing customers with potential customers. Previous research has shown that referred customers are more valuable to a firm than non-referred customers. However, previous research solely focused on the customer lifetimevalueofthenewlyreferredcustomersanddoesnotlookatthe social network characteristics. A study by Kumar et al. (2010) argues that we shouldconsider two parts of customer value, namelycustomer lifetimevalueandcustomerreferralvalue. Thelattercanbeconceived as a customer’s potential to grow the network through referrals. Early work by Granovetter (1973) highlights the importance of weak social connections, like acquaintances, in a network due to their position as bridges, connecting different communities. Extending this knowledge to customer referral programs, we can argue that referrals over weak links are powerful for accessing new communities. In this study, we investigatetheeffectofreferralsandthetiestrengthbetweentheexistingandpotentialcustomerontheresultinggrowthofthenetwork. The finding of this study are particularly useful for start-ups or marketing campaigns aiming to grow the customer base.
    • The value of neighborhood information in prospect selection models investigating the optimal level of granularity

      Van den Poel, Dirk; Baecke, Philippe (2013)
      Within analytical customer relationship management (CRM), customer acquisition models suffer the most from a lack of data quality because the information of potential customers is mostly limited to socio-demographic and lifestyle variables obtained from external data vendors. Particularly in this situation, taking advantage of the spatial correlation between customers can improve the predictive performance of these models. This study compares the predictive performance of an autoregressive and hierarchical technique in an application that identifies potential new customers for 25 products and brands. In addition, this study shows that the predictive improvement can vary significantly depending on the granularity level on which the neighborhoods are composed. Therefore, a model is introduced that simultaneously incorporates multiple levels of granularity resulting in even more accurate predictions.