A heuristic method for consumable resource allocation in multiclass dynamic PERT networks
 4.2k Downloads
 1 Citations
Abstract
This investigation presents a heuristic method for consumable resource allocation problem in multiclass dynamic Project Evaluation and Review Technique (PERT) networks, where new projects from different classes (types) arrive to system according to independent Poisson processes with different arrival rates. Each activity of any project is operated at a devoted service station located in a node of the network with exponential distribution according to its class. Indeed, each project arrives to the first service station and continues its routing according to precedence network of its class. Such system can be represented as a queuing network, while the discipline of queues is first come, first served. On the basis of presented method, a multiclass system is decomposed into several singleclass dynamic PERT networks, whereas each class is considered separately as a minisystem. In modeling of singleclass dynamic PERT network, we use Markov process and a multiobjective model investigated by Azaron and TavakkoliMoghaddam in 2007. Then, after obtaining the resources allocated to service stations in every minisystem, the final resources allocated to activities are calculated by the proposed method.
Keywords
Project management Multiclass dynamic PERT network QueuingIntroduction
Today, multiproject scheduling which considers all projects of organization as one system and therefore sharing limited resources among multiple projects is mostly noticed, whereas up to 90% of organizations handle the projects in a multiproject environment (Payne 1995). On the other hand, for better management, in some organizations, the projectzoriented approach is primarily adopted and the operations of the organization are dependently executed on projects.
In the literature of multiproject scheduling, multiproject resource constrained scheduling problem (MPRCSP) in static and deterministic conditions is a major topic of most researches. Generally, two approaches for analyzing MPRCSP exist. One approach by linking all projects of organization synthetically together into a large single project, analyzes MPRCSP, while the other approach by considering the projects as independent component and also resources constraints, proposes an objective function that contains all projects (possibly appropriately weighted) for analyzing MPRCSP.
Wiest (1967) and Pritsker et al. (1969) were the first pioneering researchers in studying MPRCSP who presented, respectively, a zero–one programming approach and a heuristic model for analyzing this problem. Then Kurtulus and Davis (1982) and Kurtulus and Narula (1985), by employing the priority rules and describing measures, studied the MPRCSP. Moreover, some investigations have been concentrated on MPRCSP by applying multiobjective and multicriteria approaches; for example, Chen (1994) proposed the zero–one goal programming model in MPRCSP for the maintenance of mineral processing, and Lova et al. (2000) analyzed MPRCSP with applying a lexicographically two criteria. Also recently, Kanagasabapathi et al. (2009) by defining performance measures and considering maximum tardiness and mean tardiness of projects studied the MPRCSP in a static environment.
On the other hand, some researchers such as Tsubakitani and Deckro (1990), Lova and Tormos (2001), Kumanan et al. (2006), Goncalves et al. (2008), Ying et al. (2009), and Chen and Shahandashti (2009) proposed heuristic and metaheuristic algorithms for solving MPRCSP. Kruger and Scholl (2008) extended the MPRCSP by considering transfer times and its cost.
In all of the mentioned investigations, MPRCSP has been analyzed in static and deterministic environment, while a few studies have been concentrated on multiproject scheduling under uncertainty conditions. FatemiGhomi and Ashjari (2002) by considering a multichannel queuing introduced a simulation model for multiproject resource allocation with stochastic activity durations. Nozick et al. (2004) also presented a nonlinear mixedinteger programming model to optimally allocate the resources, while the probability distribution of activity duration and allocated resources depend on together. Moreover, eventdriven approach and Critical Chain Project Management (CCPM) approach were proposed for overcoming uncertainty in multiproject system by Kao et al. (2006) and Byali and Kannan (2008), respectively.
Normally, greater resource allocation will incur a greater expense; therefore, we have a tradeoff between the time to complete an activity and its cost (a function of the assigned resource). When the task duration is a linear function of the allocated resource, we have the linear time–cost tradeoff problem which was analyzed by Fulkerson (1961). The nonlinear case of the time–cost tradeoff problem was studied by Elmaghraby and Salem (1982, 1984) and Elmaghraby (1993, 1996). Moreover, Elmaghraby and Morgan (2007) presented a fast method for optimizing resource allocation in a singlestage stochastic activity network.
Clearly, multiproject scheduling is more elaborate than singleproject scheduling. On the other hand, in some organizations, not only the task durations are uncertain, but also new projects dynamically enter to the organization over the time horizon. In such conditions, the organization is faced with a multiproject system in which the scheduling procedure would be more elaborate than before. Adler et al. (1995) studied such multiproject system by considering the organization as a ‘stochastic processing network’ and using simulation. They assumed that the organization is comprised of a collection of ‘service stations’ (work stations) or ‘resources’, where several types (classes) of projects exist in a system and one or more identical parallel servers can be settled in each service station. Therefore, such multiproject system, denominated as Dynamic PERT Network, can be considered as a queuing network and is attractive for organizations having the same projects, for example, maintenance projects.
AnaviIsakow and Golany (2003) by applying simulation study used the concept of CONWIP (constant workinprocess) in multiclass dynamic Project Evaluation and Review Technique (PERT) networks. They described two control mechanisms, constant number of projects in process (CONPIP) and constant time of projects in process (CONTIP), namely, CONPIP mechanism restricts the number of projects, while CONTIP mechanism limits the total processing time by all the projects that are active in the system.
In addition, Cohen et al. (2005, 2007) by using cross entropy, based on simulation, investigated the resource allocation problem in multiclass dynamic PERT networks, where the resources can work in parallel. Indeed, the number of resources allocated and servers in each service station are equal (e.g., mechanical work station with mechanics, electrical work station with electricians), and the total number of resources in the system also is constant.
On the other hand, Azaron and TavakkoliMoghaddam (2007) proposed an analytical multiobjective model to optimally control consumable resources allocated to the activities in a dynamic PERT network, where only one type of project exists in the system and the number of servers in every service station is either 1 or infinity. They assumed that the activity durations are exponentially distributed random variables, resources allocated affect the mean of service times, and the new projects are generated according to a Poisson process. A risk element was considered in a dynamic PERT network by Li and Wang (2009) who, by using general project risk element transmission theory, proposed a multiobjective risktime–cost tradeoff problem. Recently, Azaron et al. (2011) introduced an algorithm in computing optimal constant lead time for each particular project in a repetitive (dynamic) PERT network by minimizing the average aggregate cost per project.
In practice, due to various projects' requirements, most organizations execute the projects from several classes, where new projects from different classes arrive to the system dynamically over the time horizon and service stations that stochastically serve to projects. In such conditions, organization is encountered with multiclass dynamic PERT networks, while projects from different classes differ in their precedence networks, the mean of their service time in every service station, and also their arrival rates.
Reviewing the abovementioned investigations indicated that the multiclass dynamic PERT network has been studied only using simulation. As the main contribution of this paper, we present a heuristic method for the consumable resource allocation problem (or time–cost tradeoff problem) in multiclass dynamic PERT networks. Note that the resource allocation problem in multiclass dynamic PERT networks is a generalization of the resource allocation problem in dynamic PERT network, investigated by Azaron and TavakkoliMoghaddam (2007) by considering only one type (class) of projects in multiproject system.
In this paper, it is assumed that the new projects from different classes, including all their activities, are entered to the system according to independent Poisson processes with different arrival rates and each activity of any project is executed at a devoted service station settled in a node of the network based on first come, first served (FCFS) discipline. Moreover, the number of servers in every service station is either 1 or infinity, while the service times (activity durations) are independent random variables with exponential distributions. Indeed, each project arrives to the first service station and continues its routing according to precedence network of its class.
For determining the resources allocated to the service stations, the multiclass system is decomposed into several singleclass dynamic PERT networks, whereas each class is considered separately as a minisystem. For modeling of a singleclass dynamic PERT network, we use Markov process and a multiobjective model, investigated by Azaron and TavakkoliMoghaddam (2007), namely, we first convert the network of queues into a stochastic network. Then, by constructing a proper finitestate absorbing continuoustime Markov model, a system of differential equations is created. Moreover, we apply a multiobjective model containing four conflicted objectives to optimally control the consumable resources allocated to service stations in the singleclass dynamic PERT network and further employ the goal attainment method to solve a discretetime approximation of the primary multiobjective problem. Finally, after obtaining the resources allocated to service stations in every minisystem, the final resources allocated to servers are calculated by the proposed method.
This paper is composed of five sections. The remainder is organized as follows. In ‘Single class dynamic PERT networks,’ we model the singleclass dynamic PERT network by employing a finitestate continuoustime Markov process and apply a multiobjective model to optimally control the consumable resources allocated to service stations in a singleclass dynamic PERT network. In ‘Consumable resource allocation in multiclass dynamic PERT networks,’ a heuristic model is proposed for consumable resources allocated in the multiclass dynamic PERT networks. We solve an example in ‘An illustrative case’ section and conclusion is given in ‘Conclusions.’
Singleclass dynamic PERT networks
In this section, we represent a multiobjective model to optimally control the consumable resources allocated to the service stations investigated by Azaron and TavakkoliMoghaddam (2007). For this purpose, we first explain an analytical method to compute the distribution function of project completion time and then we describe a multiobjective model in singleclass dynamic PERT network.
It is assumed that a project is represented as an activityonnode (AoN) graph, and also the new projects, including all their activities, are arrived to the system according to a Poisson process with the rate of λ. Moreover, each activity of a project is executed in a devoted service station settled in a node of the network based on FCFS discipline.
Such system can be considered as a network of queues, where the service times represent the durations of the corresponding activities. It is also assumed that the number of servers in each service station to be either one or infinity, while the service times (activity durations) are independent random variables with exponential distributions.
Project completion time distribution
For presenting an analytical method to compute the distribution function for singleclass dynamic PERT network, the method of Kulkarni and Adlakha (1986) will be used because this method presents an analytical, simple, and easy approach to implement on a computer and a computationally stable algorithm to evaluate the distribution function of the project completion time. This method models PERT network with independent and exponentially distributed activity durations by continuoustime Markov chains with upper triangular generator matrices. The special structure of the chain allows us to develop very simple algorithms for the exact analysis of the network.
It is assumed that the service time in service station a is exponentially distributed with the rate of μ_{ a }, where the arrival stream of projects to each service station is according to a Poisson process with the rate of λ. Moreover, the number of server in node a is either one or infinity and therefore the node a treats as an M/M/1 or M/M/∞ model. The main steps of method to determine a finitestate absorbing continuoustime Markov process to computing the distribution function in singleclass dynamic PERT network are as follows:
Step 1. Compute the density function of the sojourn time (waiting time plus activity duration) in each service station.
Step 2. Transform the singleclass dynamic PERT network, represented as an AoN graph, to a classic PERT network represented as an activityonarc (AoA) graph.
When considering AoN graph, substitute each node with a stochastic arc (activity) whose length is equal to the sojourn time in the corresponding queuing system. For this purpose, node a in the AoN graph is replaced with a stochastic activity. Assume that b_{1},b_{2},…,b_{ n } are the incoming arcs to node a and d_{1},d_{2},…,d_{ m } are the outgoing arcs from it in the queuing network. Then, node a is substituted by activity (v,w), whose length is equal to the activity duration a. Furthermore, all arcs b_{1},b_{2},…,b_{ n } terminate to node v while all arcs d_{1},d_{2},…,d_{ m } originate from node w [for more details, see Azaron and Modarres (2005 )].
Step 3. Calculate the distribution function of the project completion time (longest path time) in the AoA obtained in step 2, while activity durations are distributed exponentially. For the computation of longest path time distribution in AoA, the method of Kulkarni and Adlakha (1986) is applied.
Let G = (V, A) be the PERT network, obtained in step 2, with a single source and a single sink, in which V represents the set of nodes and A represents the set of arcs of the stochastic network (AoA). Let s and t be the source and sink nodes, respectively, and length of arc a ϵ A be a random variable that exponentially distributed with parameter γ_{ a }. For a ϵ A, the starting node and the ending node of arc a are denoted as α(a) and β(a), respectively.
An (s,t) cut $\left(X,\stackrel{\_}{X}\right)$ is denominated a uniformly directed cut (UDC), if $\left(X,\stackrel{\_}{X}\right)=\xd8$, i.e., there are no two arcs in the cut belonging to the same path in the project network. Each UDC is clearly a set of arcs, in which the starting node of each arc belongs to X and the ending node of that arc belongs to $\stackrel{\_}{X}$.
Example 1. Consider the stochastic network shown in Figure 1 taken from Azaron and TavakkoliMoghaddam (2007). According to the definition, the UDCs of this network are (1,2); (2,3); (1,4,6); (3,4,6); and (5,6).
Definition 3. (E,F), subsets of A, is defined as admissible 2partition of a UDC D if D = E ∪ F and E ∩ F = φ, and also I(β(a)) ⊄ F for any a ∈ F.
Again consider Example 1 that (1,4,6) is one of the UDCs. For example, this cut can be decomposed into E = {1, 6} and F = {4}. In this case, the cut is an admissible 2partition because I(β(4)) = {3, 4} ⊄ F. Furthermore, if E = {6} and F = {1, 4}, then the cut is not an admissible 2partition because I(β(1)) = {1} ⊂ F = {1, 4}
 (i)
Active. An activity a is active at time t, if it is being performed at time t.
 (ii)
Dormant. An activity a is called dormant at time t, if it has completed but there is at least one unfinished activity in I(β(a)) at time t.
 (iii)
Idle. An activity a is denominated idle at time t, if it is neither active nor dormant at time t.
and X(t) = (Y(t), Z(t)).
All admissible 2partition cuts of the example network
Cut  Value 

1  (1,2) 
2  (2,3) 
3  (2,3^{a}) 
4  (1,4,6) 
5  (1,4^{a},6) 
6  (1,4,6^{a}) 
7  (1,4^{a},6^{a}) 
8  (3,4,6) 
9  (3^{a},4,6) 
10  (3,4^{a},6) 
11  (3,4,6^{a}) 
12  (3^{a},4,6^{a}) 
13  (3,4,6) 
14  (5,6) 
15  (5^{a},6) 
16  (5,6^{a}) 
17  (ø,ø) 
The set of all admissible 2partition cuts of the network is defined as S and also $\stackrel{\_}{S}=S\cup \left\{\left(\varphi ,\varphi \right)\right\}$. Note that X(t) = (ø,ø) presents that the all activities are idle at time t and therefore the project is finished by time t. It is demonstrated that {X(t), t ≥ 0} is a finitestate absorbing continuoustime Markov process (for more details, see Kulkarni and Adlakha (1986)).
A continuoustime Markov process, {X(t), t ≥ 0}, is with finitestate space $\stackrel{\_}{S}$ and since q{(ø,ø),(ø,ø)} = 0, the project is completed. In this Markov process, all of states except X(t) = (ø,ø) that is the absorbing state are transient. Furthermore, the states in $\stackrel{\_}{S}$ should be numbered such that this Q matrix be an upper triangular one. It is assumed that the states are numbered as $1,2,\dots ,N=\stackrel{\_}{\leftS\right}$ such that X(t) = (O(s), ø) and X(t) = (ø,ø) are state 1 (initial state) and state N (absorbing state), respectively.
Let T be the length of the longest path or the project completion time in the PERT network, obtained in step 2. Obviously, T = min{t ≥ 0 : X(t) = NX(0) = 1 }.
then F(t) = P_{1}(t).
where P′(t) and Q represent the derivation of the state vector P(t) and the infinitesimal generator matrix of the stochastic process {X(t), t ≥ 0}, respectively.
Multiobjective consumable resource allocation
In this section, we apply a multiobjective model to optimally control the consumable resources allocated to the service stations in a singleclass dynamic PERT network, representing as a network of queues, where we allocate more resources to service station and where the mean service time will be decreased and the direct cost of service station per period will be increased. This means the mean service time in each service station is a nonincreasing function and the direct cost of each service station per period is a nondecreasing function of the amount of resource allocated to it, i.e., the total direct costs of service station per period and the mean project completion time are dependent together and an appropriate tradeoff between them is required. Also, the variance of the project completion time should be accounted in the model because the mean and the variance are two complementary concepts. The last objective that should be considered is the probability that the project completion time does not exceed a certain threshold for ontime delivery performance.
Let U_{ a } be the maximum amount of resource available to be allocated to the activity a (a ∈ A), L_{ a } be the minimum amount of resource needed to execute the activity a, x = [x_{ a } : a ∈ A]^{ T }, and J be the amount that represents the resource available to be allocated to all of activities. Moreover, we define u as a threshold time that project completion time does not exceed it. In practice, d_{ a }(x_{ a }) and g_{ a }(x_{ a }) can be obtained by employing linear regression based on the previous similar activities or applying the judgments of experts in this area.
 1.Minimizing the project direct cost per period$\mathrm{Min}\phantom{\rule{0.25em}{0ex}}{f}_{1}\left(x\right)={\displaystyle {\sum}_{a\in A}{d}_{a}\left({x}_{a}\right)}$(12)
 2.Minimizing the mean of project completion time (P ^{′} _{1}(t) is the density function of project completion time)$\begin{array}{ll}\phantom{\rule{.5em}{0ex}}\mathrm{Min}\phantom{\rule{0.5em}{0ex}}{f}_{2}\left(x\right)& =E\left(T\right)={\displaystyle {\int}_{0}^{\infty}\left(1{P}_{1}\left(t\right)\right)\mathit{dt}}\\ ={\displaystyle {\int}_{0}^{\infty}t{{P}^{\prime}}_{1}\left(\left.t\right)\right)\mathit{dt}}\end{array}$(13)
 3.Minimizing the variance of project completion time$\begin{array}{ll}\phantom{\rule{0.7em}{0ex}}\mathrm{Min}\phantom{\rule{0.25em}{0ex}}{f}_{3}\left(x\right)& =\mathrm{Var}\left(T\right)\\ ={\displaystyle {\int}_{0}^{\infty}{t}^{2}{{P}^{\prime}}_{1}\left(\left.t\right)\right)\mathit{dt}}{\left[{\displaystyle {\int}_{0}^{\infty}t{{P}^{\prime}}_{1}\left(\left.t\right)\right)\mathit{dt}}\right]}^{2}\end{array}$(14)
 4.Maximizing the probability that the project completion time does not exceed a certain threshold$\mathrm{Max}\phantom{\rule{0.5em}{0ex}}{f}_{4}\left(x\right)={P}_{1}\left(u\right)=P\left(T\le u\right)$(15)
Goal attainment method
We now need to use a multiobjective method to solve (20), and we actually use goal attainment technique for this purpose. The goal attainment method needs to determine a goal, b_{ j }, and a weight, c_{ j }, for every objective, namely, the total project direct cost per period, the mean project completion time, the variance of project completion time, and the probability that the project completion time does not exceed a certain threshold. c_{ j }s represent the importance of objectives, whereas, if an objective has the smallest c_{ j }, then it will be the most important objective. c_{ j }s (j = 1,2,3) are commonly normalized such that $\sum _{j=1}^{3}{c}_{j}}=1$. Goal attainment method is actually a variation of goal programming method intending to minimize the maximum weighted deviation from the goals.
Consumable resource allocation in multiclass dynamic PERT networks
In practice, due to various projects' requirements, most organizations execute the projects from several classes, where new projects from different classes arrive to the system dynamically over the time horizon and service stations that stochastically serve to projects. In such conditions, organization is encountered with multiclass dynamic PERT networks, while projects from different classes differ in their precedence networks, the mean of their service time in every service station, and also their arrival rates.
In this section, we present a heuristic algorithm for consumable resource allocation problem in multiclass dynamic PERT networks, while this problem is a generalization of the resource allocation problem in dynamic PERT network, as investigated by Azaron and TavakkoliMoghaddam (2007) by considering only one type (class) of projects in multiproject system. For this purpose, it is assumed that we have K different classes from projects, where the new projects of class i (i = 1, …, K) arrive to system according to Poisson process with the rate of λ_{ i }, and the service times in service station a for the projects of class i are exponentially distributed with the rate of ${\mu}_{a}^{i}$. On the basis of this algorithm, multiclass dynamic PERT networks problem is decomposed into K singleclass dynamic PERT network problem, while each class is considered separately as a minisystem. In every minisystem i (i = 1, …, K), all projects from different classes are considered as class i with the arrival rate of $\lambda \phantom{\rule{0.5em}{0ex}}\left(={\displaystyle \sum _{i=1}^{K}{\lambda}_{i}}\right)$. Then every minisystem is solved by using the steps found in the sections of ‘Single class dynamic PERT networks’ as a singleclass dynamic PERT network problem.
Consequently, the proposed algorithm is as follows.
Proposed Algorithm:
Step 1. Convert the multiclass dynamic PERT networks problem into K singleclass dynamic PERT network problem (minisystem). For this purpose, in every minisystem i, consider the all projects from different classes as class i with the arrival rate of $\lambda \left(\phantom{\rule{0.5em}{0ex}}={\displaystyle \sum _{i=1}^{K}{\lambda}_{i}}\right)$.
Step 2. For every minisystem i, compute the resources allocated to service stations, x^{ i } = [x_{ a }^{ i }:a ∊ A]^{ T }, and objective function, z^{ i }, by the model represented in ‘Multiobjective consumable resource allocation’ section.
Step 3. Calculate the resources allocated to service stations, x = [x_{ a }:a ∊ A]^{ T }, by $x={\displaystyle \sum _{i=1}^{K}\frac{{\lambda}_{i}}{\lambda}}{x}^{i}$ and also obtain the objective function of multiclass system by $z={\displaystyle \sum _{i=1}^{K}\frac{{\lambda}_{i}}{\lambda}}{z}^{i}$ .
An illustrative case
In this section, to illustrate the proposed heuristic algorithm, we consider the three classes of projects depicted in Figure 2. The assumptions are as follows:

The system has three different classes of projects.

The new projects from different classes, including all their activities, are generated according to Poisson processes with the rates of λ_{1} = 2.5, λ_{2} = 1.5, λ_{3} = 1 per year.

The activity durations in service station a for the projects of class i (i = 1, 2, 3) are exponentially distributed. Table 2 shows the characteristics of the activities in different classes, where the time unit and the cost unit are, respectively, in year and in thousand dollars.

There are infinite servers in service station 3 and other stations have only one server settled in the nodes.

The capacity of the system is infinite.

The threshold time that the project completion time does not exceed it for every minisystem is 3.

The maximum amount of consumable resource available to be allocated to all service stations is J = 15.

In all experiments, the value of ε is equal 0.01.

The goals and the weights of objectives for goal attainment method in every class, are b_{1} = 25, b_{2} = 1, b_{3} = 0.25, b_{4} = 0.95 and c_{1} = 0.3, c_{2} = 0.1, c_{3} = 0.3, c_{4} = 0.3, respectively. Note that the goals and the weights of objectives for different classes can be different.
Characteristics of the activities
Activity (a)  d_{ a }(x_{ a })  Project type  g_{ a }(x_{ a })  L _{ a }  U _{ a } 

1  1.2x_{1} + 1  1  \0.15 − 0.02x_{1}  1  5 
2  0.17 − 0.02x_{1}  
3  0.12 − 0.015x_{1}  
2  1.2x_{2} + 0.8  1  0.2 − 0.015x_{2}  1  6 
2  0.18 − 0.01x_{2}  
3  0.15 − 0.01x_{2}  
3  1.8x_{3} + 1  1  0.8 − 0.1x_{3}  1  5 
2  0.9 − 0.11x_{3}  
3  0.85 − 0.1x_{3}  
4  x_{4} + 0.7  1  0.12 − 0.01x_{4}  1  5 
2  0.1 − 0.01x_{4}  
3  0.14 − 0.02x_{4}  
5  1.3x_{5} + 1.5  1  0.14 − 0.02x_{5}  1  6 
2  0.12 − 0.01x_{5}  
3  0.15 − 0.02x_{5}  
6  1.5x_{6} + 0.5  1  0.18 − 0.01x_{6}  1  5 
2  0.16 − 0.01x_{6}  
3  0.2 − 0.02x_{6} 
In this example, threeclass dynamic PERT networks problem is decomposed into three singleclass dynamic PERT network problem, while each class is considered separately as a minisystem. In every minisystem i(i = 1,2,3), all projects from different classes are considered as class i with the arrival rate of λ = 5(=2.5 + 1.5 + 1). Then, every minisystem is solved as singleclass dynamic PERT network problem by using the steps found in the sections of ‘Single class dynamic PERT networks.’
The computational results
Project type  R value  Δt  x _{ 1 }  x _{ 2 }  x _{ 3 }  x _{ 4 }  x _{ 5 }  x _{ 6 }  z  f _{ 1 }  f _{ 2 }  f _{ 3 }  f _{ 4 }  CT 

1  50  0.16  1.732  4.312  1.143  1  1.893  4.920  7.332  25.651  1.733  0.399  0.935  00′:07″ 
80  0.1  1.735  4.313  1.132  1  1.896  4.923  7.331  25.646  1.733  0.538  0.931  00′:25″  
100  0.08  1.737  4.314  1.126  1  1.898  4.925  7.331  25.64  1.733  0.590  0.931  00′:31″  
2  50  0.16  2.619  4.685  3.056  1  1  2.639  5.435  26.025  1.543  0.271  0.972  00′:04″ 
80  0.1  2.874  4.379  2.411  1  1  3.335  5.342  25.847  1.534  0.377  0.970  00′:08″  
100  0.08  2.875  4.381  2.408  1  1  3.336  5.342  25.846  1.534  0.408  0.965  00′:21″  
3  50  0.16  1  2.545  2.070  2.019  3.056  4.218  2.665  25.800  1.267  0.218  0.983  00′:10″ 
80  0.1  1  2.541  1.887  1.988  3.366  4.192  2.656  25.797  1.266  0.308  0.982  00′:19″  
100  0.08  1  2.542  1.886  1.988  3.367  4.192  2.656  25.797  1.266  0.333  0.980  00′:35″ 
To illustrate the performance of proposed algorithm, we evaluate its results with a simulation method that randomly allocate the resources to service stations and simulate the multiclass dynamic PERT network through Mont Carlo simulation. Note that Fatemi Ghomi and Hashemin (1999) proposed conditional Monte Carlo simulation and crude Monte Carlo simulation to compute the network completion time distribution function. However, the simulation method will be as follows:
Simulation method:
Step 1. n=1
Step 2. Generate one initial solution x = [x_{ a } : a ∈ A]^{ T }, randomly, where x_{ a } ∈ [L_{ a }, U_{ a }] and $\sum _{a\in A}{x}_{a}}\le J$
Step 3. Simulate the multiclass dynamic PERT network (Figure 2) according to the initial solution
Step 4. Calculate the objective function, z, for the initial solution
Step 5. Repeat

Generate a new solution, ${x}^{\mathrm{new}}={\left[{x}_{a}^{\mathrm{new}}:a\in A\right]}^{T}$, randomly, where ${x}_{a}^{\mathrm{new}}\in \left[{L}_{a},{U}_{a}\right]$ and $\sum _{a\in A}{x}_{a}}\le J$

Simulate the multiclass dynamic PERT network according to the new solution

Calculate the objective function, z^{new}, for the new solution

Calculate Δz = z^{new} − z

If ∆z≺0 then x = x^{new}, z = z^{new}
 $n=n+1$
Until n ≥ N, where N is number of simulation run
Step 6. Present x and z
Conclusions
In this article, we proposed a heuristic method for the consumable resource allocation problem (or time–cost trade off problem) in multiclass dynamic PERT networks. It was assumed that the new projects of different classes, including all their activities, are entered to the system according to an independent Poisson processes, and each activity of any project is executed at a devoted service station settled in a node of the network according to its class. The number of servers in every service station is either 1 or infinity, while the service times (activity durations) are independent random variables with exponential distributions. Indeed, each project arrives to the first service station and continues its routing according to its precedence network of corresponding class. Such system was considered as a queuing network, while the discipline of queues is first come, first served.
For determining the resources allocated to the service stations, the multiclass system was decomposed into several singleclass dynamic PERT networks, whereas each class was considered separately as a minisystem. For modeling of singleclass dynamic PERT network, we used Markov process and a multiobjective model, investigated by Azaron and TavakkoliMoghaddam (2007), namely, we first converted the network of queues into a stochastic network. Then, by constructing a proper finitestate absorbing continuoustime Markov model, a system of differential equations was created. Next, we applied a multiobjective model containing four conflicted objectives to optimally control the resources allocated to service stations in singleclass dynamic PERT network and further used the goal attainment method to solve a discretetime approximation of the primary multiobjective problem. Finally, after obtaining the resources allocated to service stations in every minisystem, the final resources allocated to servers were calculated by proposed algorithm.
In multiobjective model, the total project direct cost was considered as an objective to be minimized and the mean project completion time, as another effective objective, should also be accounted that to be minimized. The variance of the project completion time was another effective objective in the model because the mean and the variance are two complementary concepts. The probability that the project completion time does not exceed a certain threshold was considered as the last objective.
For obtaining the best optimal allocated consumable resource in singleclass dynamic PERT network, we considered the various combinations of portions for specific time interval. Based on the presented example, If the length of every portion is decreased, the accuracy of solution is increased, i.e., the value of objective is decreased and the computational time, CT, is also increased.
Authors' information
SY is an Assistant Professor of Industrial Engineering at Iran University of Science and Technology (IUST), Tehran, Iran. His current research interests include stochastic processes and their applications, project scheduling and supply chains. He has contributed articles to different international journals, such as European Journal of Operational Research, International Journal of System Science, Iranian Journal of Science & Technology, and International Journal of Advanced Manufacturing Technology. SN is the Associate Professor of Industrial Engineering Department in IUST since October 1990. The teaching experiences are mainly in project management, human resource management, and engineering economics. MMM also is the Assistant Professor of Industrial Engineering Department in IUST. His research interests include performance measurement systems, inventory planning, and control and strategic planning.
Notes
Supplementary material
References
 1.Adler PS, Mandelbaum A, Nguyen V, Schwerer E: From project to process management: an empiricallybased framework for analyzing product development time. Manag Sci 1995,41(3):458–484. 10.1287/mnsc.41.3.458zbMATHCrossRefGoogle Scholar
 2.AnaviIsakow S, Golany B: Managing multiproject environments through constant workinprocess. Int J Proj Manag 2003,21(1):9–18. 10.1016/S02637863(01)000588CrossRefGoogle Scholar
 3.Azaron A, Modarres M: Distribution function of the shortest path in networks of queues. OR Spectr 2005, 27: 123–144. 10.1007/s0029100401693zbMATHMathSciNetCrossRefGoogle Scholar
 4.Azaron A, TavakkoliMoghaddam R: Multiobjective time–cost tradeoff in dynamic PERT networks using an interactive approach. Eur J Oper Res 2007, 180: 1186–1200. 10.1016/j.ejor.2006.05.014zbMATHCrossRefGoogle Scholar
 5.Azaron A, Fynes B, Modarres M: Due date assignment in repetitive projects. Int J Prod Econ 2011, 129: 79–85. 10.1016/j.ijpe.2010.09.004CrossRefGoogle Scholar
 6.Byali RP, Kannan MV: Critical Chain Project Management  a new project management philosophy for multi project environment. J Spacecraft Technol 2008, 18: 30–36.Google Scholar
 7.Chen V: A 0–1 goal programmingmodel for scheduling multiple maintenance projects at a copper mine. Eur J Oper Res 1994, 6: 176–191.CrossRefGoogle Scholar
 8.Chen PH, Shahandashti SM: Hybrid of genetic algorithm and simulated annealing for multiple project scheduling with multiple resource constraints. Autom Constr 2009, 18: 434–443. 10.1016/j.autcon.2008.10.007CrossRefGoogle Scholar
 9.Cohen I, Golany B, Shtub A: Managing stochastic, finite capacity, multiproject systems through the cross entropy methodology. Ann Oper Res 2005, 134: 183–199. 10.1007/s1047900557301zbMATHMathSciNetCrossRefGoogle Scholar
 10.Cohen I, Golany B, Shtub A: Resource allocation in stochastic, finitecapacity, multiproject systems through the cross entropy methodology. J Sched 2007, 10: 181–193. 10.1007/s1095100700130zbMATHMathSciNetCrossRefGoogle Scholar
 11.Elmaghraby SE: Resource allocation via dynamic programming in activity networks. Eur J Oper Res 1993, 64: 199–215. 10.1016/03772217(93)90177OzbMATHCrossRefGoogle Scholar
 12.Elmaghraby SE: Optimal procedures for the discrete time/cost tradeoff problem in project networks. Eur J Oper Res 1996, 88: 50–68. 10.1016/03772217(94)001812zbMATHCrossRefGoogle Scholar
 13.Elmaghraby SE, Morgan CD: A samplepath optimization approach for optimal resource allocation in stochastic projects. J Econ Manag 2007, 3: 367–389.Google Scholar
 14.Elmaghraby SE, Salem AM: Optimal project compression under convex functions, I and II. Appl Manag Sci 1982, 2: 1–39.Google Scholar
 15.Elmaghraby SE, Salem AM: Optimal linear approximation in project compression. IIE Trans 1984, 16: 339–347. 10.1080/07408178408975253CrossRefGoogle Scholar
 16.Fatemi Ghomi SMT, Ashjari B: A simulation model for multiproject resource allocation. Int J Proj Manag 2002, 20: 127–130. 10.1016/S02637863(00)000387CrossRefGoogle Scholar
 17.Fatemi Ghomi SMT, Hashemin SS: A new analytical algorithm and generation of Gaussian quadrature formula for stochastic network. Eur J Oper Res 1999, 114: 610–625. 10.1016/S03772217(98)001970zbMATHCrossRefGoogle Scholar
 18.Fulkerson DR: A network flow computation for project curves. Manag Sci 1961, 7: 167–178. 10.1287/mnsc.7.2.167zbMATHMathSciNetCrossRefGoogle Scholar
 19.Goncalves JF, Mendes JJM, Resende MGC: A genetic algorithm for the resource constrained multiproject scheduling problem. Eur J Oper Res 2008, 189: 1171–1190. 10.1016/j.ejor.2006.06.074zbMATHCrossRefGoogle Scholar
 20.Kanagasabapathi B, Rajendran C, Ananthanarayanan K: Performance analysis of scheduling rules in resourceconstrained multiple projects. Int J Ind Syst Eng 2009, 4: 502–535.Google Scholar
 21.Kao HP, Hsieh B, Yeh Y: A petrinet based approach for scheduling and rescheduling resourceconstrained multiple projects. J Chinese Inst Ind Eng 2006,23(6):468–477.Google Scholar
 22.Kruger D, Scholl A: Managing and modelling general resource transfers in (multi) project scheduling. OR Spectr 2008, 32: 369–394.MathSciNetCrossRefGoogle Scholar
 23.Kulkarni V, Adlakha V: Markov and Markovregenerative PERT networks. Oper Res 1986, 34: 769–781. 10.1287/opre.34.5.769zbMATHMathSciNetCrossRefGoogle Scholar
 24.Kumanan S, Jegan JG, Raja K: Multiproject scheduling using an heuristic and a genetic algorithm. Int J Adv Manuf Technol 2006, 31: 360–366. 10.1007/s0017000501992CrossRefGoogle Scholar
 25.Kurtulus IS, Davis EW: Scheduling: categorization of heuristic rules performance. Manag Sci 1982, 28: 161–172. 10.1287/mnsc.28.2.161zbMATHCrossRefGoogle Scholar
 26.Kurtulus IS, Narula SC: Multiproject scheduling: analysis of project performance. IIE Trans 1985, 17: 58–66. 10.1080/07408178508975272CrossRefGoogle Scholar
 27.Li C, Wang K: The risk element transmission theory research of multiobjective risktime–cost tradeoff. Compute Math with Appl 2009, 57: 1792–1799. 10.1016/j.camwa.2008.10.051zbMATHCrossRefGoogle Scholar
 28.Lova A, Tormos P: Analysis of scheduling schemes and heuristic rules performance in resourceconstrained multiproject scheduling. Ann Oper Res 2001, 102: 263–286. 10.1023/A:1010966401888zbMATHMathSciNetCrossRefGoogle Scholar
 29.Lova A, Maroto C, Tormos P: A multicriteria heuristic method to improve resource allocation in multiproject scheduling. Eur J Oper Res 2000, 127: 408–424. 10.1016/S03772217(99)004907zbMATHCrossRefGoogle Scholar
 30.Nozick LK, Turnquist MA, Xu N: Managing portfolios of projects under uncertainty. Ann Operation Res 2004, 132: 243–256.zbMATHMathSciNetCrossRefGoogle Scholar
 31.Payne JH: Management of multiple simultaneous projects: a stateoftheart review. Int J Proj Manag 1995, 13: 163–168. 10.1016/02637863(94)000199CrossRefGoogle Scholar
 32.Pritsker AAB, Watters LJ, Wolfe PM: Multiproject scheduling with limited resources: a zero–one programming approach. Manag Sci 1969,16(1):93–108. 10.1287/mnsc.16.1.93CrossRefGoogle Scholar
 33.Tsubakitani S, Deckro RF: A heuristic for multiproject scheduling with limited resources in the housing industry. Eur J Oper Res 1990, 49: 80–91. 10.1016/03772217(90)90122RCrossRefGoogle Scholar
 34.Wiest JD: A heuristic model for scheduling large projects with limited resources. Manag Sci 1967,13(6):359–377.CrossRefGoogle Scholar
 35.Ying Y, Shou Y, Li M: Hybrid genetic algorithm for resource constrained multiproject scheduling problem. J Zhejiang University (Eng Sci) 2009, 43: 23–27.zbMATHGoogle Scholar
Copyright information
This article is published under license to BioMed Central Ltd. This is an Open Access article distributed under the terms of the Creative Commons Attribution License (http://creativecommons.org/licenses/by/2.0), which permits unrestricted use, distribution, and reproduction in any medium, provided the original work is properly cited.