Abstract Increasing the success rate of a process, i.e. the percentage of cases that end in a positive outcome, is a recurrent process improvement goal. At runtime, there are often certain actions (a.k.a. treatments) that workers may execute to lift the probability that a case ends in a positive outcome. For example, in a loan origination process, a possible treatment is to issue multiple loan offers to increase the probability that the customer takes a loan. Each treatment has a cost. Thus, when defining policies for prescribing treatments to cases, managers need to consider the net gain of the treatments. Also, the effect of a treatment varies over time: treating a case earlier may be more effective than later in a case. This paper presents a prescriptive monitoring method that automates this decision-making task. The method combines causal inference and reinforcement learning to learn treatment policies that maximize the net gain. The method leverages a conformal prediction technique to speed up the convergence of the reinforcement learning mechanism by separating cases that are likely to end up in a positive or negative outcome, from uncertain cases. An evaluation on two real-life datasets shows that the proposed method outperforms a state-of-the-art baseline.

Abstract Predictive process monitoring is a family of techniques to analyze events produced during the execution of a business process in order to predict the future state or the final outcome of running process instances. Existing techniques in this field are able to predict, at each step of a process instance, the likelihood that it will lead to an undesired outcome. These techniques, however, focus on generating predictions and do not prescribe when and how process workers should intervene to decrease the cost of undesired outcomes. This paper proposes a framework for prescriptive process monitoring, which extends predictive monitoring with the ability to generate alarms that trigger interventions to prevent an undesired outcome or mitigate its effect. The framework incorporates a parameterized cost model to assess the cost–benefit trade-off of generating alarms. We show how to optimize the generation of alarms given an event log of past process executions and a set of cost model parameters. The proposed approaches are empirically evaluated using a range of real-life event logs. The experimental results show that the net cost of undesired outcomes can be minimized by changing the threshold for generating alarms, as the process instance progresses. Moreover, introducing delays for triggering alarms, instead of triggering them as soon as the probability of an undesired outcome exceeds a threshold, leads to lower net costs.

Reducing cycle time is a recurrent concern in the field of business process management. Depending on the process, various interventions may be triggered to reduce the cycle time of a case, for example, using a faster shipping service in an order-to-delivery process or calling a customer to obtain missing information rather than waiting passively. However, each of these interventions comes with a cost. This paper tackles the problem of determining if and when to trigger a time-reducing intervention in a way that maximizes a net gain function. The paper proposes a prescriptive monitoring method that uses orthogonal random forests to estimate the causal effect of triggering a time-reducing intervention for each ongoing case of a process. Based on this estimate, the method triggers interventions according to a user-defined policy. The method is evaluated on two real-life datasets.

Personalization is one of the key applications in machine learning with widespread usage across e-commerce, entertainment, production, healthcare and many other industries. While various machine learning techniques present novel state-of-the-art advances and super-human performance year-over-year, personalization and recommender-systems applications are often late-adopters of novel solutions due to problem hardness and implementation complexity. This tutorial presents recent advances across the personalization industry and demonstrates their practical applications in real case-studies of world-leading online platforms. Key trends such as deep learning, causality and active exploration with bandits are depicted with real examples and demonstrated alongside their business considerations and implementation challenges.Rising topics like explainability, fairness, natural interfaces and content generation are covered, touching on aspects of both technology and user experience. Our tutorial relies on recent advances in the field and on work conducted at Booking.com, where we implement personalization models on one of the world's leading online travel platform.

This paper proposes an approach to analyze an event log of a business process in order to generate case-level recommendations of treatments that maximize the probability of a given outcome. Users classify the attributes in the event log into controllable and non-controllable, where the former correspond to attributes that can be altered during an execution of the process (the possible treatments). We use an action rule mining technique to identify treatments that co-occur with the outcome under some conditions. Since action rules are generated based on correlation rather than causation, we then use a causal machine learning technique, specifically uplift trees, to discover subgroups of cases for which a treatment has a high causal effect on the outcome after adjusting for confounding variables. We test the relevance of this approach using an event log of a loan application process and compare our findings with recommendations manually produced by process mining experts.

Predictive business process monitoring methods exploit historical process execution logs to generate predictions about running instances (called cases) of a business process, such as the prediction of the outcome, next activity, or remaining cycle time of a given process case. These insights could be used to support operational managers in taking remedial actions as business processes unfold, e.g., shifting resources from one case onto another to ensure the latter is completed on time. A number of methods to tackle the remaining cycle time prediction problem have been proposed in the literature. However, due to differences in their experimental setup, choice of datasets, evaluation measures, and baselines, the relative merits of each method remain unclear. This article presents a systematic literature review and taxonomy of methods for remaining time prediction in the context of business processes, as well as a cross-benchmark comparison of 16 such methods based on 17 real-life datasets originating from different industry domains.

Predictive business process monitoring refers to the act of making predictions about the future state of ongoing cases of a business process, based on their incomplete execution traces and logs of historical (completed) traces. Motivated by the increasingly pervasive availability of fine-grained event data about business process executions, the problem of predictive process monitoring has received substantial attention in the past years. In particular, a considerable number of methods have been put forward to address the problem of outcome-oriented predictive process monitoring, which refers to classifying each ongoing case of a process according to a given set of possible categorical outcomes—e.g., Will the customer complain or not? Will an order be delivered, canceled, or withdrawn? Unfortunately, different authors have used different datasets, experimental settings, evaluation measures, and baselines to assess their proposals, resulting in poor comparability and an unclear picture of the relative merits and applicability of different methods. To address this gap, this article presents a systematic review and taxonomy of outcome-oriented predictive process monitoring methods, and a comparative experimental evaluation of eleven representative methods using a benchmark covering 24 predictive process monitoring tasks based on nine real-life event logs.

The enactment of business processes is generally supported by information systems that record data about each process execution (a.k.a. case). This data can be analyzed via a family of methods broadly known as process mining. Predictive process monitoring is a process mining technique concerned with predicting how running (uncompleted) cases will unfold up to their completion. In this paper, we propose a predictive process monitoring framework for estimating the probability that a given predicate will be fulfilled upon completion of a running case. The framework takes into account both the sequence of events observed in the current trace, as well as data attributes associated to these events. The prediction problem is approached in two phases. First, prefixes of previous (completed) cases are clustered according to control flow information. Second, a classifier is built for each cluster using event data attributes to discriminate between cases that lead to a fulfillment of the predicate under examination and cases that lead to a violation within the cluster. At runtime, a prediction is made on a running case by mapping it to a cluster and applying the corresponding classifier. The framework has been implemented in the ProM toolset and validated on a log pertaining to the treatment of cancer patients in a large hospital.

A key problem for facilitators of online communication and social networks is to identify users whose activity is likely to change in the near future. Such predictions may serve as basis for targeted campaigns aimed at sustaining or increasing overall user engagement in the network. A common approach to this problem is to apply machine learning methods to make predictions at the level of individuals. These approaches consider only information about each individual user and, thus, do not exploit the social connections and structure of the network. In this paper, we approach the problem of activity change prediction at the level of communities rather than individuals. We develop predictive models of activity change over communities obtained using state-of-art community detection methods and compare their predictive power with each other and against the single-user baseline and ego networks. The results show that community-level prediction models achieve higher prediction accuracy than the traditional single-user approach, whereas a local community detection algorithm outperforms a global modularity-based method.