We envision a new logistic process for delivering parcels by crowdsourcing the tasks of the couri... more We envision a new logistic process for delivering parcels by crowdsourcing the tasks of the couriers. It is based on a network of automatic service points, which are used as a drop-off, pickup, and intermediate transfer points. The system offers the couriers monetary rewards for stopping by the service points and for transferring parcels between them during their regular trips. We present a stochastic dynamic program that yields an optimal parcel routing policy in polynomial time under some simplifying assumptions. In particular, this policy ignores the capacity constraints of the service points and the couriers. A heuristic policy that is based on the value table of the above dynamic program is devised to relax the simplifying assumptions and adapt it to realistic settings. The economic viability of the proposed method is demonstrated by an extensive simulation study that is based on realistic data of car journeys and small parcel shipments in a metropolitan area. Our experiment shows that even if a small fraction of the drivers is willing to participate, the average delivery time is few hours. Moreover, the rewards paid to the occasional couriers relatively to their engagement time is above the average hourly wage while the average cost of delivering a parcel is significantly lower than the price of parcel delivery service.
The classic test collection problem (TCP) selects a minimal set of binary tests needed to classif... more The classic test collection problem (TCP) selects a minimal set of binary tests needed to classify the state of a system correctly. The TCP has applications in various domains, such as the design of monitoring systems in engineering, communication, and healthcare. In this paper, we define the stochastic test collection problem (STCP) that generalizes the TCP. While the TCP assumes that the tests' results can be deterministically mapped into classes, in the STCP, the results are mapped to probability distributions over the classes. Moreover, each test and each type of classification error is associated with some cost. A solution of the STCP is a subset of tests and a mapping of their results to classes. The objective is to minimize the weighted sum of the tests' costs and the expected cost of the classification errors. We present an integer linear programming formulation of the problem and solve it using a commercial solver. To solve larger instances, we apply three metaheuristics for the STCP, namely, Tabu Search (TS), Cross-Entropy (CE), and Binary Gravitational Search Algorithm (BGSA). These methods are tested on publicly available datasets and shown to deliver nearly optimal solutions in a fraction of the time required for the exact solution.
Bike sharing systems (BSSs) allow customers to rent bicycles at automatic rental stations distrib... more Bike sharing systems (BSSs) allow customers to rent bicycles at automatic rental stations distributed throughout a city, use them for a short period of time, and return them to any station. One of the major issues that BSS operators must address is nonhomogeneous asymmetric demand processes. These demand processes create an inherent imbalance, thus leading to shortages either of bicycles when users are attempting to rent them and of vacant lockers when users are attempting to return them. The predominant approach taken by operators to cope with this difficulty is to reposition bicycles to rebalance the inventory levels at the different stations. Most repositioning studies assume that a target inventory level or range of inventory levels is known for each station. In this paper, we focus on determining the correct target level for repositioning according to a well-defined objective. This is a challenging task because of the nature of the user behavior that creates the interactions am...
EURO Journal on Transportation and Logistics, 2019
In this study, we introduce a logistic model for the delivery of small parcels to a set of servic... more In this study, we introduce a logistic model for the delivery of small parcels to a set of service points (SPs), and we present effective methods for solving it. In the traditional delivery model, each recipient specifies a single location at which they wish to receive the parcel; however, when SPs are used, many recipients may have no strong preference among several locations, e.g., near the recipient's home address, near the recipient's office, or in the recipient's favorite shopping mall. If some recipients are flexible and willing to provide the sender with more than one delivery location, it is possible to perform the delivery task at lower cost and within a shorter amount of time. Our solution methods are based on the concepts of the savings heuristic, the petal method and tabu search with a large neighborhood. An extensive numerical study is conducted to evaluate our solution methods and demonstrate the benefits of our model compared to the traditional nonflexible one. We also present a simulation study to demonstrate that our model can be adapted to a stochastic and dynamic environment.
In this paper, we study a single-vehicle routing problem with stochastic service times, stochasti... more In this paper, we study a single-vehicle routing problem with stochastic service times, stochastic time-dependent travel times, and soft time windows, where the travel times or the service times may be inter-dependent. The objective is to minimize the expected length of the working day plus penalties for late arrivals. The stochasticity is modeled using a set of scenarios based on historical data. This approach enables the spatial and temporal interdependencies in the road network to be captured. We introduce a specialized branch-andbound algorithm and a successful tabu search heuristic for the problem. In an extensive numerical experiment based on real historical travel time data, we demonstrate the applicability of both methods to problem instances of up to 18 customers and 40 scenarios. These dimensions are a safe upper bound for instances originating from the domain of field service operation. The resulting routes are tested on realistic scenarios that were not included in the i...
We present a model for policy optimization of the online booking of mobile personnel over a multi... more We present a model for policy optimization of the online booking of mobile personnel over a multiday horizon with a different cutoff for each day, where the goal is to maximize the expected ratio of accepted requests at steady-state. This model fits the practice of many service providers who allow booking of time slots over a horizon of multiple days and use availability control of the demand. Since the planning horizon is indefinite and the service horizon of each day overlaps the horizon of subsequent days, the objective is defined in terms of the steadystate performance. The interactions with the customers are performed in a single step: The system offers an assortment of time slots covering the next few days, and the user either chooses one of them or abandons the system. Upon the arrival of a service request, the provider estimates the opportunity cost of serving the request at each of the available time slots. We model this cost as a linear function and a Cobb–Douglas function...
The well-studied Test Collection Problem (TCP) selects a minimal set of binary tests needed to co... more The well-studied Test Collection Problem (TCP) selects a minimal set of binary tests needed to correctly classify the state of a system. This model has applications in various domains, such as the design of monitoring systems in engineering, communication and healthcare. In this paper, we define the Stochastic Test Collection Problem (STCP) that generalizes the TCP. While the original TCP assumes that each combination of the outputs of the tests can be mapped into a class, in the STCP, these combinations are mapped to probability distributions over the classes. Moreover, each test and each type of classification error is associated with some cost. The objective is to select a subset of tests that minimize the weighted sum of the tests’ costs and the expected cost of the classification errors. The STCP further generalizes the TCP by allowing general categorical results of the tests rather than binary ones. We apply three metaheuristic methods for the STCP, namely, Tabu Search (TS), C...
Following a recent interest in sustainable scheduling under operational costs that vary over time... more Following a recent interest in sustainable scheduling under operational costs that vary over time, we study scheduling problems on a single machine under time-of-use electricity tariffs. We consider two main variants of the problem: cost minimization and profit maximization. In the cost minimization problem, the set of jobs to be processed is given, and the goal is to schedule all jobs within a planning horizon so as to minimize the total cost, while in the profit maximization problem, one needs to select a set of jobs to be processed such that the total profit is maximized. The general cases of the cost minimization and profit maximization problems in which preemptions are forbidden are strongly NP-hard. In this paper, we show that some special cases with identical processing times can be solved by efficient algorithms. In addition, we consider several extensions of the problems, including release times, due dates, and variable energy consumption.
In complex systems, such as automobiles, aircraft, and even smart cities, it is possible to insta... more In complex systems, such as automobiles, aircraft, and even smart cities, it is possible to install a large number of sensors that can be used to monitor their states and make decisions regarding their operation and maintenance. However, because the sensors bear their costs and increase the complexity of the monitoring system, it is desirable to select the most inexpensive set of sensors that is sufficient to provide accurate and reliable information. In this paper, we introduce the sensor selection problem and formulate it as an integer linear program. The model is an extension of the minimum test collection problem with respect to several constraints, such as the sensors’ cost which is not necessarily equal. We present an effective exact solution method that uses the special structure of the integer-programming model to reduce its dimension so it can easily be solved. This solution method was implemented and demonstrated to be superior in comparison to a state-of-the-art integer p...
In this paper, we study a single-vehicle routing problem with stochastic service times, stochasti... more In this paper, we study a single-vehicle routing problem with stochastic service times, stochastic time-dependent travel times, and soft time windows, where the travel times may be interdependent. The objective is to minimize the expected length of the route duration plus penalties for late arrivals. The stochasticity is modeled using a set of scenarios based on historical data. This approach enables the spatial and temporal interdependencies in the road network to be captured. We introduce a specialized branch-and-bound algorithm and a successful adaptive large neighborhood search heuristic for the problem. In a numerical experiment based on real historical travel time data, we demonstrate the applicability of both methods to problem instances of up to 36 customers and 40 scenarios. These dimensions are a safe upper bound for instances originating from the field service operation domain. The resulting routes are tested on realistic scenarios that were not included in the problem input (the training set) to demonstrate the merits of using historical data. Compared with solutions that ignore the time dependency and/or stochasticity of the parameters, our solutions are consistently superior.
In this paper, the constraint programming (CP) approach is applied for the simple assembly line b... more In this paper, the constraint programming (CP) approach is applied for the simple assembly line balancing problem (SALBP) as well as some of its generalizations. CP is a rich modeling language that enables the formulation of general combinatorial problems and is coupled with a strong set of solution methods that are available through general purpose solvers. The proposed formulations are conversions of well-known mixed integer programming (MILP) formulations to CP, along with a new set of constraints that helps the CP solver to converge faster. As a generic solution method, we compare its performance with the best known generic MILP formulations and show that it consistently outperforms MILP for medium to large problem instances. A comparison with SALOME, a well-known custom-made algorithm for solving the SALBP-1, shows that both approaches are capable of efficiently solving problems with up to 100 tasks. When 1000-task problems are concerned, SALOME provides better performance; however, CP can provide relatively good close to optimal solutions for multiple combinations of problem parameters. Finally, the generality of the CP approach is demonstrated by some adaptations of the proposed formulation to other variants of the assembly line balancing problem including the U-shaped assembly line balancing problem and the task assignment and equipment selection problem.
Quality and Reliability Engineering International, 2016
This work proposes a method to improve the QoS provided to internet users by website servers. In ... more This work proposes a method to improve the QoS provided to internet users by website servers. In particular, the goal is to minimize the expected number of browsing steps (clicks), also known as the expected path length, which are required to reach a website page by a community of users. We use Markov chain modeling to represent the transition probabilities from one webpage to another, and the first passage from page to page that can be retrieved from web server logs. The proposed method deletes links among webpages to minimize the expected path length of the website. Three different methods for achieving this goal are examined: (i) a greedy deletion heuristic; (ii) an approximated branch and bound algorithm; and (iii) a cross-entropy metaheuristic. Numerical studies show that the proposed greedy heuristic results in the optimal solution in more than 60% of the tested cases, while in almost 90% of the cases, the obtained solution is within 10% of the optimal solution. Both the approximated branch and bound and the cross-entropy methods achieved optimality in more than 80% of the tested cases; however, this came with a much higher computational cost.
EURO Journal on Transportation and Logistics, 2015
An increased use of shared mobility systems such as bike sharing, ride sharing or car sharing is ... more An increased use of shared mobility systems such as bike sharing, ride sharing or car sharing is evident in recent years. These systems gain popularity as an alternative to using private vehicles since they have the advantage of reducing traffic congestion, parking space shortage, and air pollution. Shared mobility systems serve as a good complementary to mass transit systems, for example, as a convenient solution to the first and last miles of a journey. In addition, members of vehicle sharing programs have a lower tendency to own a private car. Designing and operating vehicle sharing systems raises many interesting challenges. These problems range from long-term strategic issues to on-line operational decisions. On the strategic end of the spectrum, there are questions such as the economical, technological, and environmental viability of deploying new car or bike-sharing systems in a city. There are various possible modes of operation of vehicle sharing system: it can be station-based or free-float, round-trip or one-way (A to B). Next, the stations' location and capacity should be determined and at a more tactical level, pricing incentives and reservation policies should be set. Finally, at the operational level, vehicles' and stations' maintenance as well as repositioning/ rebalancing decisions should be taken on a daily basis and sometimes in real time. Various tools may be used to solve the above problems: forecasting using statistical and data mining methods, stochastic modeling, simulation and optimization using exact and heuristic methods. Vehicle sharing systems use modern & Tal Raviv
It has been a long (and fruitful) journey that could not have ended this way without your support... more It has been a long (and fruitful) journey that could not have ended this way without your support, guidance and insights. I have certainly learned a great deal about the nature of academic research during our mutual work and for that-I will be eternally grateful. I wish to thank the employees of ClickSoftware for their willingness to talk about this research, its methodology and findings. In particular, I wish to express a special gratitude to Israel Beniaminy, Senior Vice President of Product Strategy in ClickSoftware for his insights and advices. These were especially valuable. I wish to thank my Mother, Malka, and my siblings, Vered and Gil, for their endless support and encouragement throughout this long process. Last, but definitely not least, I wish to express my deepest gratitude to my father, Sahilu Avraham. As long as I can remember, my education has been his prime focus and this achievement is his no less than it is mine.
13th IFAC Symposium on Information Control Problems in Manufacturing, 2009
Supply Chain Management (SCM) is concerned with decisions related to the flow of materials throug... more Supply Chain Management (SCM) is concerned with decisions related to the flow of materials through the value chain of specific goods. From its inception, and by most definitions, SCM is attempting to integrate decisions related to procurement of raw materials, manufacturing, warehousing and distribution of the finished products to the final customers. However, the large integrated problem is highly intractable. We propose to investigate functional decomposition methods of a supply chain problem of quite a general structure. In these decompositions, decisions related to one or more functions (for example, distribution, procurement, production, and inventories) are made separately and subsequently glued together to create an integrated solution. We formulate the general problem, which includes several important extensions to existing literature, then develop and compare various functional decompositions. The comparison is along solution times, as well as optimality gaps, compared to the integrated problem. We obtain insights with respect to the preferred decomposition method and its organizational implications.
We envision a new logistic process for delivering parcels by crowdsourcing the tasks of the couri... more We envision a new logistic process for delivering parcels by crowdsourcing the tasks of the couriers. It is based on a network of automatic service points, which are used as a drop-off, pickup, and intermediate transfer points. The system offers the couriers monetary rewards for stopping by the service points and for transferring parcels between them during their regular trips. We present a stochastic dynamic program that yields an optimal parcel routing policy in polynomial time under some simplifying assumptions. In particular, this policy ignores the capacity constraints of the service points and the couriers. A heuristic policy that is based on the value table of the above dynamic program is devised to relax the simplifying assumptions and adapt it to realistic settings. The economic viability of the proposed method is demonstrated by an extensive simulation study that is based on realistic data of car journeys and small parcel shipments in a metropolitan area. Our experiment shows that even if a small fraction of the drivers is willing to participate, the average delivery time is few hours. Moreover, the rewards paid to the occasional couriers relatively to their engagement time is above the average hourly wage while the average cost of delivering a parcel is significantly lower than the price of parcel delivery service.
The classic test collection problem (TCP) selects a minimal set of binary tests needed to classif... more The classic test collection problem (TCP) selects a minimal set of binary tests needed to classify the state of a system correctly. The TCP has applications in various domains, such as the design of monitoring systems in engineering, communication, and healthcare. In this paper, we define the stochastic test collection problem (STCP) that generalizes the TCP. While the TCP assumes that the tests' results can be deterministically mapped into classes, in the STCP, the results are mapped to probability distributions over the classes. Moreover, each test and each type of classification error is associated with some cost. A solution of the STCP is a subset of tests and a mapping of their results to classes. The objective is to minimize the weighted sum of the tests' costs and the expected cost of the classification errors. We present an integer linear programming formulation of the problem and solve it using a commercial solver. To solve larger instances, we apply three metaheuristics for the STCP, namely, Tabu Search (TS), Cross-Entropy (CE), and Binary Gravitational Search Algorithm (BGSA). These methods are tested on publicly available datasets and shown to deliver nearly optimal solutions in a fraction of the time required for the exact solution.
Bike sharing systems (BSSs) allow customers to rent bicycles at automatic rental stations distrib... more Bike sharing systems (BSSs) allow customers to rent bicycles at automatic rental stations distributed throughout a city, use them for a short period of time, and return them to any station. One of the major issues that BSS operators must address is nonhomogeneous asymmetric demand processes. These demand processes create an inherent imbalance, thus leading to shortages either of bicycles when users are attempting to rent them and of vacant lockers when users are attempting to return them. The predominant approach taken by operators to cope with this difficulty is to reposition bicycles to rebalance the inventory levels at the different stations. Most repositioning studies assume that a target inventory level or range of inventory levels is known for each station. In this paper, we focus on determining the correct target level for repositioning according to a well-defined objective. This is a challenging task because of the nature of the user behavior that creates the interactions am...
EURO Journal on Transportation and Logistics, 2019
In this study, we introduce a logistic model for the delivery of small parcels to a set of servic... more In this study, we introduce a logistic model for the delivery of small parcels to a set of service points (SPs), and we present effective methods for solving it. In the traditional delivery model, each recipient specifies a single location at which they wish to receive the parcel; however, when SPs are used, many recipients may have no strong preference among several locations, e.g., near the recipient's home address, near the recipient's office, or in the recipient's favorite shopping mall. If some recipients are flexible and willing to provide the sender with more than one delivery location, it is possible to perform the delivery task at lower cost and within a shorter amount of time. Our solution methods are based on the concepts of the savings heuristic, the petal method and tabu search with a large neighborhood. An extensive numerical study is conducted to evaluate our solution methods and demonstrate the benefits of our model compared to the traditional nonflexible one. We also present a simulation study to demonstrate that our model can be adapted to a stochastic and dynamic environment.
In this paper, we study a single-vehicle routing problem with stochastic service times, stochasti... more In this paper, we study a single-vehicle routing problem with stochastic service times, stochastic time-dependent travel times, and soft time windows, where the travel times or the service times may be inter-dependent. The objective is to minimize the expected length of the working day plus penalties for late arrivals. The stochasticity is modeled using a set of scenarios based on historical data. This approach enables the spatial and temporal interdependencies in the road network to be captured. We introduce a specialized branch-andbound algorithm and a successful tabu search heuristic for the problem. In an extensive numerical experiment based on real historical travel time data, we demonstrate the applicability of both methods to problem instances of up to 18 customers and 40 scenarios. These dimensions are a safe upper bound for instances originating from the domain of field service operation. The resulting routes are tested on realistic scenarios that were not included in the i...
We present a model for policy optimization of the online booking of mobile personnel over a multi... more We present a model for policy optimization of the online booking of mobile personnel over a multiday horizon with a different cutoff for each day, where the goal is to maximize the expected ratio of accepted requests at steady-state. This model fits the practice of many service providers who allow booking of time slots over a horizon of multiple days and use availability control of the demand. Since the planning horizon is indefinite and the service horizon of each day overlaps the horizon of subsequent days, the objective is defined in terms of the steadystate performance. The interactions with the customers are performed in a single step: The system offers an assortment of time slots covering the next few days, and the user either chooses one of them or abandons the system. Upon the arrival of a service request, the provider estimates the opportunity cost of serving the request at each of the available time slots. We model this cost as a linear function and a Cobb–Douglas function...
The well-studied Test Collection Problem (TCP) selects a minimal set of binary tests needed to co... more The well-studied Test Collection Problem (TCP) selects a minimal set of binary tests needed to correctly classify the state of a system. This model has applications in various domains, such as the design of monitoring systems in engineering, communication and healthcare. In this paper, we define the Stochastic Test Collection Problem (STCP) that generalizes the TCP. While the original TCP assumes that each combination of the outputs of the tests can be mapped into a class, in the STCP, these combinations are mapped to probability distributions over the classes. Moreover, each test and each type of classification error is associated with some cost. The objective is to select a subset of tests that minimize the weighted sum of the tests’ costs and the expected cost of the classification errors. The STCP further generalizes the TCP by allowing general categorical results of the tests rather than binary ones. We apply three metaheuristic methods for the STCP, namely, Tabu Search (TS), C...
Following a recent interest in sustainable scheduling under operational costs that vary over time... more Following a recent interest in sustainable scheduling under operational costs that vary over time, we study scheduling problems on a single machine under time-of-use electricity tariffs. We consider two main variants of the problem: cost minimization and profit maximization. In the cost minimization problem, the set of jobs to be processed is given, and the goal is to schedule all jobs within a planning horizon so as to minimize the total cost, while in the profit maximization problem, one needs to select a set of jobs to be processed such that the total profit is maximized. The general cases of the cost minimization and profit maximization problems in which preemptions are forbidden are strongly NP-hard. In this paper, we show that some special cases with identical processing times can be solved by efficient algorithms. In addition, we consider several extensions of the problems, including release times, due dates, and variable energy consumption.
In complex systems, such as automobiles, aircraft, and even smart cities, it is possible to insta... more In complex systems, such as automobiles, aircraft, and even smart cities, it is possible to install a large number of sensors that can be used to monitor their states and make decisions regarding their operation and maintenance. However, because the sensors bear their costs and increase the complexity of the monitoring system, it is desirable to select the most inexpensive set of sensors that is sufficient to provide accurate and reliable information. In this paper, we introduce the sensor selection problem and formulate it as an integer linear program. The model is an extension of the minimum test collection problem with respect to several constraints, such as the sensors’ cost which is not necessarily equal. We present an effective exact solution method that uses the special structure of the integer-programming model to reduce its dimension so it can easily be solved. This solution method was implemented and demonstrated to be superior in comparison to a state-of-the-art integer p...
In this paper, we study a single-vehicle routing problem with stochastic service times, stochasti... more In this paper, we study a single-vehicle routing problem with stochastic service times, stochastic time-dependent travel times, and soft time windows, where the travel times may be interdependent. The objective is to minimize the expected length of the route duration plus penalties for late arrivals. The stochasticity is modeled using a set of scenarios based on historical data. This approach enables the spatial and temporal interdependencies in the road network to be captured. We introduce a specialized branch-and-bound algorithm and a successful adaptive large neighborhood search heuristic for the problem. In a numerical experiment based on real historical travel time data, we demonstrate the applicability of both methods to problem instances of up to 36 customers and 40 scenarios. These dimensions are a safe upper bound for instances originating from the field service operation domain. The resulting routes are tested on realistic scenarios that were not included in the problem input (the training set) to demonstrate the merits of using historical data. Compared with solutions that ignore the time dependency and/or stochasticity of the parameters, our solutions are consistently superior.
In this paper, the constraint programming (CP) approach is applied for the simple assembly line b... more In this paper, the constraint programming (CP) approach is applied for the simple assembly line balancing problem (SALBP) as well as some of its generalizations. CP is a rich modeling language that enables the formulation of general combinatorial problems and is coupled with a strong set of solution methods that are available through general purpose solvers. The proposed formulations are conversions of well-known mixed integer programming (MILP) formulations to CP, along with a new set of constraints that helps the CP solver to converge faster. As a generic solution method, we compare its performance with the best known generic MILP formulations and show that it consistently outperforms MILP for medium to large problem instances. A comparison with SALOME, a well-known custom-made algorithm for solving the SALBP-1, shows that both approaches are capable of efficiently solving problems with up to 100 tasks. When 1000-task problems are concerned, SALOME provides better performance; however, CP can provide relatively good close to optimal solutions for multiple combinations of problem parameters. Finally, the generality of the CP approach is demonstrated by some adaptations of the proposed formulation to other variants of the assembly line balancing problem including the U-shaped assembly line balancing problem and the task assignment and equipment selection problem.
Quality and Reliability Engineering International, 2016
This work proposes a method to improve the QoS provided to internet users by website servers. In ... more This work proposes a method to improve the QoS provided to internet users by website servers. In particular, the goal is to minimize the expected number of browsing steps (clicks), also known as the expected path length, which are required to reach a website page by a community of users. We use Markov chain modeling to represent the transition probabilities from one webpage to another, and the first passage from page to page that can be retrieved from web server logs. The proposed method deletes links among webpages to minimize the expected path length of the website. Three different methods for achieving this goal are examined: (i) a greedy deletion heuristic; (ii) an approximated branch and bound algorithm; and (iii) a cross-entropy metaheuristic. Numerical studies show that the proposed greedy heuristic results in the optimal solution in more than 60% of the tested cases, while in almost 90% of the cases, the obtained solution is within 10% of the optimal solution. Both the approximated branch and bound and the cross-entropy methods achieved optimality in more than 80% of the tested cases; however, this came with a much higher computational cost.
EURO Journal on Transportation and Logistics, 2015
An increased use of shared mobility systems such as bike sharing, ride sharing or car sharing is ... more An increased use of shared mobility systems such as bike sharing, ride sharing or car sharing is evident in recent years. These systems gain popularity as an alternative to using private vehicles since they have the advantage of reducing traffic congestion, parking space shortage, and air pollution. Shared mobility systems serve as a good complementary to mass transit systems, for example, as a convenient solution to the first and last miles of a journey. In addition, members of vehicle sharing programs have a lower tendency to own a private car. Designing and operating vehicle sharing systems raises many interesting challenges. These problems range from long-term strategic issues to on-line operational decisions. On the strategic end of the spectrum, there are questions such as the economical, technological, and environmental viability of deploying new car or bike-sharing systems in a city. There are various possible modes of operation of vehicle sharing system: it can be station-based or free-float, round-trip or one-way (A to B). Next, the stations' location and capacity should be determined and at a more tactical level, pricing incentives and reservation policies should be set. Finally, at the operational level, vehicles' and stations' maintenance as well as repositioning/ rebalancing decisions should be taken on a daily basis and sometimes in real time. Various tools may be used to solve the above problems: forecasting using statistical and data mining methods, stochastic modeling, simulation and optimization using exact and heuristic methods. Vehicle sharing systems use modern & Tal Raviv
It has been a long (and fruitful) journey that could not have ended this way without your support... more It has been a long (and fruitful) journey that could not have ended this way without your support, guidance and insights. I have certainly learned a great deal about the nature of academic research during our mutual work and for that-I will be eternally grateful. I wish to thank the employees of ClickSoftware for their willingness to talk about this research, its methodology and findings. In particular, I wish to express a special gratitude to Israel Beniaminy, Senior Vice President of Product Strategy in ClickSoftware for his insights and advices. These were especially valuable. I wish to thank my Mother, Malka, and my siblings, Vered and Gil, for their endless support and encouragement throughout this long process. Last, but definitely not least, I wish to express my deepest gratitude to my father, Sahilu Avraham. As long as I can remember, my education has been his prime focus and this achievement is his no less than it is mine.
13th IFAC Symposium on Information Control Problems in Manufacturing, 2009
Supply Chain Management (SCM) is concerned with decisions related to the flow of materials throug... more Supply Chain Management (SCM) is concerned with decisions related to the flow of materials through the value chain of specific goods. From its inception, and by most definitions, SCM is attempting to integrate decisions related to procurement of raw materials, manufacturing, warehousing and distribution of the finished products to the final customers. However, the large integrated problem is highly intractable. We propose to investigate functional decomposition methods of a supply chain problem of quite a general structure. In these decompositions, decisions related to one or more functions (for example, distribution, procurement, production, and inventories) are made separately and subsequently glued together to create an integrated solution. We formulate the general problem, which includes several important extensions to existing literature, then develop and compare various functional decompositions. The comparison is along solution times, as well as optimality gaps, compared to the integrated problem. We obtain insights with respect to the preferred decomposition method and its organizational implications.
We envision a new logistic process for delivering parcels by crowdsourcing the tasks of the couri... more We envision a new logistic process for delivering parcels by crowdsourcing the tasks of the couriers. It is based on a network of automatic service points, which are used as a drop-off, pickup, and intermediate transfer points. The system offers the couriers monetary rewards for stopping by the service points and for transferring parcels between them during their regular trips. We present a stochastic dynamic program that yields an optimal parcel routing policy in polynomial time under some simplifying assumptions. In particular, this policy ignores the capacity constraints of the service points and the couriers. A heuristic policy that is based on the value table of the above dynamic program is devised to relax the simplifying assumptions and adapt it to realistic settings. The economic viability of the proposed method is demonstrated by an extensive simulation study that is based on realistic data of car journeys and small parcel shipments in a metropolitan area. Our experiment shows that even if a small fraction of the drivers is willing to participate, the average delivery time is few hours. Moreover, the rewards paid to the occasional couriers relatively to their engagement time is above the average hourly wage while the average cost of delivering a parcel is significantly lower than the price of parcel delivery service.
Uploads
Papers by Tal Raviv