@incollection{AbeParRes99a,
	author = "J. Abello and P.M. Pardalos and M.G.C. Resende",
	title = "On maximum clique problems in very large graphs",
	booktitle = "External memory algorithms and visualization",
	editor = "J. Abello and J. Vitter",
        series = "{DIMACS} Series on Discrete Mathematics and Theoretical
                  Computer Science",
	publisher = "American Mathematical Society",
	volume = "50",
	pages = "119--130",
	year = "1999",
        annote   =	"An approach for clique and quasi-clique
                	computations in very large multi-digraphs is presented.
                	The authors discuss graph decomposition schemes
                	that break up the original problem into several
                	pieces of manageable dimensions. A semi-external
                	memory {GRASP}
                	is presented to approximately solve the maximum clique
                	problem and maximum quasi-clique problem.
                	The construction phase uses vertex degrees
                	as a guide for construction. The local search uses a
                	simple $(2,1)$-exchange."
}

@Article{AbeResSud02a,
        author = {J. Abello and M.G.C. Resende and S. Sudarsky},
        title = {Massive quasi-clique detection},
        journal = {Lecture Notes in Computer Science},
	volume = {2286},
	pages = {598--612},
	annote = {Detecting dense subgraphs (quasi-cliques) in massive 
		sparse graphs whose vertex set fits in RAM is a hard 
		problem. 
		In this paper, a {GRASP} is designed for extracting 
		dense subgraphs on case of both nonbipartite and 
		bipartite case. 
		In case of nonbipartite graphs, the construction 
		procedure applies a greedy criterion based on 
		a measure of the effect of elements selection on 
		the potential of the other vertices. 
		In the bipartite case, the authors adapt the above 
		greedy criterion to a slight different definition 
		of potential. 
		In both cases, the neighborhood structure used 
		is a $(2,1)$-exchange.},
	year = {2002}
}

@techreport{AlvGonDeA03a,
        author = {A.M. \'Alvarez and J. L. Gonz\'alez and K. De-Alba},
        title = {Scatter search for a network design problem},
	number = {PISIS-2003-02},
	institution = {Universidad Aut\'onoma de Nuevo Le\'on,
		       Facultad de Ingenier\'{\i}a Mec\'anica y El\'ectrica,
	               Divisi\'on de Posgrado en Ingenier\'{\i}a 
		       de Sistemas, Mexico.},
	annote = {The authors propose a Scatter Search (SS) algorithm 
		for solving a fixed charge capacitated multicommodity 
		network design problems on undirected networks. 
		SS is an evolutionary method that constructs new solutions 
		by combining others. 
		It performs the following three basic steps: 1) generation of 
		a population $P$; 2) extraction of a reference set $R$; 
		3) combination of elements from $R$ and updating of $R$. 
		The different ways SS can be implemented consists in designing 
		several different subroutines, including a Diversification 
		Generation Method (DGM). 
		The DGM proposed in this paper is a {GRASP} with some 
		memory features incorpored. 
		More in detail, in the {GRASP} DGM for each commodity 
		a certain number $q$ of shortest paths between each 
		origin-destination pair are kept as RCL elements. 
		The local search consists basically in sorting the chosen 
		paths and possibly exchange some of them in order to get 
		a better distribution.},
	year = {2003}
}

@article{AmaCapMal03a,
        author = {E. Amaldi and A. Capone and F. Malucelli},
        title = {{Planning {UMTS} base station location: {O}ptimization models 
	          with power control and algorithms}},
        journal = {IEEE Transactions on Wireless Communications},
	volume = {2},
	number = {5},
	year = {2003},
	pages = {939--952},
	annote = "The UMTS base station location problem is NP-hard and 
		to solve it, the authors propose two greedy randomized 
		algorithms and a tabu search. 
		The greedy function used takes into account the fraction 
		of traffic covered and the installation costs. 
		A swap procedure is used until no improvement can be done."
}

@inproceedings{AmaCapMalSig03a,
        author = {E. Amaldi and A. Capone and F. Malucelli and F. Signori},
        title = {{Optimization models and algorithms for downlink {UMTS}
	         radio planning}},
        booktitle = {{Wireless Communications and Networking, 2003 
	              (WCNC 2003)}},
	volume = {2},
        month = {March},
	annote = "In this paper, the authors deal with the UMTS base 
                station location problem. 
                Two mathematical programming formulations are provided 
                for locating directive base stations considering downlink 
                (base station to mobile) direction and assuming a power-based 
                as well or a SIR-based (Signal-to-Interface Ratio) 
                power control mechanism. 
                Already in Amaldi et al. (2003), the authors proposed 
                two randomized heuristics, which are adapted in this paper 
                for solving the simailar problem.",
	year = {2003},
	pages = {827--831}
}

@Article{ArmKliLusRos00a,
	author = "M. Armony and J.G. Klincewicz and H. Luss and
                  M.B. Rosenwein",
	title = "Design of stacked self-healing rings using a genetic
                 algorithm",
	journal = "Journal of Heuristics",
	volume = "6",
        pages = "85--105",
	year = "2000",
        annote   =     "A genetic algorithm for design of
                    	stacked self-healing rings is proposed.
                    	The objective is to optimize the trade-off between
                    	the cost of connecting nodes to the ring and
                    	the cost of routing demand on multiple rings.
                    	The initial population of the genetic algorithm is
                    	made up of randomly generated solutions as well
                    	as solutions generated by a {GRASP}.
                    	Computational comparisons are made with a commercial
                    	integer programming package.",
}

@techreport{BruBat01a,
	author = "M. Brunato and R. Battiti",
	title = "A multistart randomized greedy algorithm for traffic 
		grooming on mesh logical topologies",
	institution = "Department of Mathematics,
			University of Trento",
	address = "Trento, Italy",
	year = "2001",
	annote = "The authors addresse a logical topology design problem on 
		Dense Wavelength Division Multiplexing (DWDM) optical 
		networks, where the traffic is measured at sub-wavelength 
		resultion and the key factor to determine the fitness of a 
		solution is the number of lightpaths required. 
		They describe a {GRASP}-like heuristic for minimizing 
		the number of lightpaths. 
		The greedy choice takes into account the load associated 
		with each lightpath, i.e. an integer value representing 
		the number of traffic unit routed to that lightpath."
}

@Article{CanResRib01a,
        author = "S.A. Canuto and M.G.C. Resende and C.C. Ribeiro",
        title = "Local search with perturbation for the prize-collecting
                 {Steiner} tree problems in graphs",
        journal = "Networks",
        volume = "38",
        pages = "50--58",
        year = "2001",
        annote   = "Several local search strategies have been investiged 
		in this paper for solving the prize-collecting {Steiner} 
		tree problems in graphs, including path relinking, VNS, 
		and a {GRASP} that uses some cost perturbations. 
		The greedy choice takes into account the input edge weights, 
		while the local search routine defines as neighborhood 
		of a solution $T(X)$ the set of all minimum spanning 
		trees $T(X')$, where $X'$ differs from $X$ for exactly 
		one node.",
}

@techreport{CooKwoGloSchTayWit02a,
        author = "B.P. Cooke and D. Kwon and D. Glotov and
                  S. Schurr and D. Taylor and T. Wittman",
        title = "Mobility management in cellular telephony",
        institution = "Institute of Mathematics and its Applications,
                       University of Minnesota, USA",
        year = "2002",
	annote = "This paper deals with the problem of minimizing 
		the cost of transferring calls from transceiver to 
		transceiver and from controller to controller. 
		Four different algorithms are proposed in this paper 
		for solving the problem: a branch-and-bound, 
		a metropolis algorithm with annealing, a 
		genetic algorithm, and a greedy search. 
		The last one has been derived from a {GRASP} 
		proposed for the quadratic assignment problem."
}

@article{GenPotSor00a,
        author = "B. Gendron and J.-Y. Potvin and P. Soriano",
        title = "Diversification strategies in local search for a 
		 nonbifurcated network loading problem",
	journal = "European Journal of Operational Research",
	volume = {142},
	number = {2},
	pages = {231--241},
	year = {2002},
	annote = 	"In this paper, a heuristic approach is proposed that
                        alternates construction phases and local search phases.
                        In the beginning, a construction method provides an
                        initial feasible solution, while at subsequent
                        construction steps a diversification approach is
                        adopted for exploiting information gathered along
                        previous iterations.
                        Two local searches are used: a pure descent search
                        and a tabu search.
                        As further improvement of their method, the
                        authors propose to implement the {GRASP} choosing
                        criterion that at random selects the next element among
                        a list of better candidates.",
}

@Article{Kli92a,
        author = "J.G. Klincewicz",
        title = "Avoiding local optima in the $p$-hub location
	         problem using tabu search and {GRASP}",
        journal = "Annals of Operations Research",
        volume = "40",
        pages = "283--302",
        year = "1992",
        annote   =     "This paper proposes two heuristics based on
                	tabu search and {GRASP} for the $p$-hub
                	location problem.  
			The objective is to overcome the
                	difficulty that local search algorithms encounter.
                	The local search procedure
                	of proposed {GRASP} is based on the $2$-exchange.",
}

@Article{Kli02a,
	 author = "J.G. Klincewicz",
	 title  = "Enumeration and search procedures for a hub location
                   problem with economies of scale",
        journal = "Annals of Operations Research",
        volume = "110",
	pages = "107--122",
	year = "2002",
        annote   =     "An optimal enumeration scheme, as well as other 
			heuristics based on tabu search and {GRASP} are 
			proposed for locating hubs in a communications or 
			transportation network. 
			The {GRASP} greedy function takes into account 
			the amount of originating and terminating traffic. 
			The local search uses a $1-1$ swap neighborhood 
			structure.",
}

@inproceedings{LiCheYin00a,
        author = "B. Li and F. Chen and L. Yin",
        title = "Server replication and its placement for reliable 
		multicast",
        booktitle = "Proceedings of the IEEE ICCCN-00",
	city = "Las Vegas",
        month = "October",
        year = "2000",
	annote = "In a multicast network, packets are forwarded from a source 
		(server) to group of receivers along a distribution tree, 
		where the source is the root, the receivers are the leaves, 
		and the multicast-capable routers are the internal nodes. 
		The problem studied in this paper consists of placing 
		multiple replicated servers within the multicast-capable 
		routers and to solve it, the authors propose a number of 
		heuristic algorithms, including a {GRASP}. 
		The greedy function is the router cost function, while 
		the local search phase uses a $k$-exchange neighborhood 
		structure with $k=1$.",
        pages = "396--401"
}

@incollection{LiuParRajRes00a,
	author = "X. Liu and P.M. Pardalos and S. Rajasekaran and
		  M.G.C. Resende",
        title = "A {GRASP} for frequency assignment in mobile radio networks",
        booktitle = "Mobile Networks and Computing",
        editor = "S. Rajasekaran and P.M. Pardalos and F.Hsu",
        series = "{DIMACS} Series on Discrete Mathematics and Theoretical
                   Computer Science",
        publisher = "American Mathematical Society",
        volume = "52",
        pages = "195--201",
        year = "2000",
        annote = 	"A {GRASP} for frequency assignment is described.
                	Local search uses simulated annealing.
                	The construction phase uses two greedy functions. The first
                	chooses a vertex from
                	the set of unselected vertices with high saturation degrees.
                	The second function is used to assign a frequency to the
                	selected vertex. A frequency is selected from a set of
                	permissible frequencies that contribute little additional
                	cost to the objective function."
}

@Article{MahRib02a, 
         author = "P. Mahey and C.C. Ribeiro",
         title  = "Modeling modern multimedia traffic",
        journal = "Annals of Operations Research",
        volume = "110",
        pages = "107--122",
        year = "2002",
        annote   =     "This paper surveys state-of-the-art algorithms 
			solving optimal routing problems on multi-service 
			communication problems.",
}

@inproceedings{Mys01a,
        author = "A. Myslek",
        title = "Greedy randomised adaptive search procedures ({GRASP}) 
		for topological design of {MPLS} networks",
        booktitle = "Proceedings of the 8th Polish Teletraffic Symposium",
        city = "Zakopane, Poland",
        year = "2001",
	annote = "The paper addresses the IP/MPLS network 
		cost optimization problem of selecting localization 
		of nodes and links, combined with link dimensioning. 
		A {GRASP} is proposed, whose greedy function uses demand 
		flow allocation, while at each iteration of the local 
		search phase, some selected nodes (or edges) become 
		unavailable if provided and vice versa."
}

@inproceedings{MysKar02a,
        author = "A. Myslek and P. Kara\'{s}",
        title = "Heuristic methods for topological design of telecomminication 
		networks",
        booktitle = "Proceedings of PGTS 2002",
	annote = "Given a list of potential node locations and a list 
		of feasible interconnections between nodes, the generic 
		topological network design problem consists 
		in finding a network structure and a demand alloction 
		pattern that minimizes the cost of the network. 
		A pool of heuristics are proposed for solving the problem, 
		including a Simulated Allocation (SAL) and a hybrid {GRASP}, 
		that uses SAL as local search. 
		SAL works with partial allocation states. 
		At each iteration, a randomly chosen demand is allocated 
		or deallocated and allocation is chosen with higher 
		probability. 
		The algorithm stops after a certain predefinite number of 
		not improving iterations.",
        year = "2002" 
}

@techreport{OliGom99a,
	author = "C.A.S. Oliveira and F.C. Gomes",
	title = "Two metaheuristics for channel allocation in mobile
		   telephony",
	institution = "Artificial Intelligence Laboratory,
			Universidade Federal do Cear\'a",
	address = "Fortaleza, Brazil",
	month = "August",
	year = "1999",
	annote =	"The frequency assignment problem
                	described consists in minimizing the total system 
			interference in mobile phone covered areas,
                	with respect to co-channel and adjacent-channel 
			interference.
                	Two metaheuristics are proposed: {GRASP} and 
			Asynchronous Team (A-Team). 
			The construction phase of the proposed {GRASP} is
                	realized by a procedure that at each step chooses the 
			next antenna to which a frequency will be assigned. 
			In the RCL construction, priority is given to 
			transmitters with fewer options of frequency 
			assignment. To implement the local search phase, 
			the authors use a down hill
                	algorithm, that performs random perturbations in the
                	solution, exchanging the frequency of one antenna by 
			another randomly chosen. 
			The stopping criterion used for the
                	down hill algorithm is execution time."
}

@techreport{Pas98a,
	 author = "E.L. Pasiliao",
	 title = "A greedy randomized adaptive search procedure
                  for the multi-criteria radio link frequency
                  assignment problem",
	 institution = "Department of ISE,
			University of Florida",
	 address = "Gainesville, FL 32611-6595",
	 year = "1998",
        annote   =      "A {GRASP} for computing approximate solutions to a 
			radio link frequency assignment problem is proposed.
                	The objective is to minimize the order and the span of 
			the solution set.
                	The local search procedure attempts to
                	eliminate each channel from the communication network.
                	This process is repeated until an attempt to eliminate
                	every frequency in the solution set has been made 
			without success."
}

@article{PinPlaCamMar04a,
        author = "E. Pi{\~n}ana and I. Plana and V. Campos and R. Mart\'{\i}",
        title = "{GRASP} and path relinking for the matrix bandwidth
                 minimization",
        journal = "European J. of Operational Research",
        volume = "153",
	number = "1",
        pages = "200--210",
        year = "2004",
        annote   =     {Given matrix $A=\{a_{ij}\}_{n\times n}$, 
			the matrix bandwidth minimization problem consists 
			in finding a permutation of the rows and columns 
			of $A$ that keeps the nonzero elements in a band 
			lying as close as possible to the main diagonal 
			of $A$. 
			The problem can be easily stated in terms of 
			graph optimization problem, considering a vertex 
			for each row and a vertex for each column and  
			connecting vertex $i$ to vertex $j$ through an edge 
			either if $a_{ij}\not= 0$ or $a_{ji}\not= 0$. 
			Then, the original problem is equivalent to the 
			problem of finding a labeling $f$ of the vertices 
			that minimizes the maximum difference between 
			labels of adjacent vertices. 
			The authors propose a {GRASP}, in which 
			the greedy choice favors vertices having a maximum 
			number of adjacent labeled vertices. 
			The local search phase is based on swap moves that 
			exchange labels of a pair of vertices.},
}

@inproceedings{PopPicAriDem97a,
	author = "F. Poppe and M. Pickavet and P. Arijs and P. Demeester",
	title = "Design techniques for {SDH} mesh-restorable networks",
	booktitle = "Proceedings of the European Conference on Networks
                     and Optical Communications (NOC'97), Volume 2:
                     Core and ATM Networks",
        pages = "94--101",
	year = "1997",
	annote =  	"To design low cost reliable
                	telecommunication networks, the authors propose
                	three algorithms: an integer
                	linear programming algorithm (branch-and-cut-and-price),
                	a {GRASP}, and a zoom-in approach that combines a 
			genetic algorithm with deterministic optimization 
			routines.
                	The greedy choice of the proposed {GRASP} is to favor
                	paths having lowest additional cost. The local search
                	iteratively tries to reroute some paths in order to 
			further decrease the overall network cost."
}

@Article{PraRib00a,
	 author = "M. Prais and C.C. Ribeiro",
	 title = "Reactive {GRASP}: {A}n application to a matrix
                  decomposition problem in {TDMA} traffic assignment",
	 journal = "INFORMS Journal on Computing",
	 volume = "12",
	 pages = "164--176",
	 year = "2000",
        annote   =     "This paper describes a {GRASP} for matrix decomposition
                	problem arising in the context of traffic assignment
                	in communication satellites. A geostationary communication
                	satellite has a number of spot beam antennas covering
                	geographically distributed areas. According to the slot
                	switching configuration on the on-board switch, the uplink
                	traffic received at the satellite has to be immediately sent
                	to ground areas through a set of transponders. The slot
                	switching configurations are determined through the solution
                	of a time slot assignment problem, which is equivalent to
                	the decomposition of a nonnegative traffic matrix into
                	the sum of a family of switching mode matrices.
			A refinement of {GRASP}, called Reactive {GRASP},
                	is proposed.
                	Instead of using a fixed value for the basic
                	parameter that defines the restrictiveness
                	of the candidate list during the construction phase,
                	Reactive {GRASP} self-adjusts the parameter value
                	according to the quality of the solutions previously found. 
			The local search phase of the {GRASP} proposed is based on
                	a new neighborhood, defined by constructive and
                	destructive moves.",
}

@inproceedings{ResRes99a,
	author = "L.I.P. Resende and M.G.C. Resende",
	title = "A {GRASP} for frame relay {PVC} routing",
        booktitle = "Proc. of the Third Metaheuristics International
                      Conference",
        city= "Angra dos Reis, Brazil",
        month = "July",
        year = "1999",
        pages = "397--402",
	annote   =	"This paper describes a {GRASP} for routing permanent virtual
                	circuits (PVC) for frame relay in telecommunications
                	systems. The objective is to minimize PVC delays while
                	balancing trunk loads. The greedy choice selects from the
                	set of not yet routed PVCs the one that minimizes the delay
                	while balancing the trunk loads. The local search procedure
                	reroutes each PVC, one at a time, checking each time if the
                	new route taken together with the remaining fixed routes
                	improves the objective function."
}

@Article{Res98a,
	author = "M.G.C. Resende",
	title = "Computing approximate solutions of the
		 maximum covering problem using {GRASP}",
	journal = "Journal of Heuristics",
	volume = "4",
        pages = "161--171",
	year = "1998",
        annote   =      "A {GRASP} for the maximum covering problem
                	is described.
                	The greedy function is the total weight of the 
			yet-uncovered demand points that become covered after 
			the selection.
                	The local search procedures uses a $2$-exchange 
			neighborhood structure.
                	The {GRASP} is shown to find near optimal solutions.",
}

@techreport{ResUlu97a,
	author = "M.G.C. Resende and O. Ulular",
	title = "{SMART}: {A} tool for {AT\&T} {W}orldnet access design --
                 {L}ocation of {C}ascade 9000 concentrators",
	institution = "AT\&T Labs Research",
	address = "Florham Park, NJ 07932 USA",
	year = "1997",
	annote = 	"This report describes {SMART}, a software tool
                	for finding low cost configurations of Cascade 9000
                	concentrators in the {AT\&T} {W}orldnet backbone access
                	network. The concentrator location problem is stated and
                	cost model is presented for concentrator configurations.
                	This cost model is used in a {GRASP}, proposed for 
			finding approximate solutions to the concentrator 
			location problem.
                	The greedy choice favors the points-of-presence (POPs)
                	with smallest incremental cost. 
			The local search implements a simple $2$-exchange."
}

@Article{RibRos02a,
        author = "C.C. Ribeiro and I. Rosseti",
        title = "A parallel {GRASP} for the 2-path network design problem",
        journal = "Lecture Notes in Computer Science",
        volume = "2004",
        pages = "922--926",
        year = "2002",
        annote   =     "Given a connected undirected graph $G(V,E)$, a 
			nonnegative edge weight function 
			$w:\ E\rightarrow {\mathbb R}^+$, 
			and a set $D$ of pairs of origin-destination nodes, the 
			$2$-path network design problem consists in finding a 
			minimum weighted subset of edges $E'\subseteq E$ 
			containing a $2$-path between the extremities of every 
			origin-destination pair in $D$. 
			In this paper, a parallel {GRASP} with path-relinking 
			is proposed for solving this problem.",
}

@phdthesis{Ros03a,
        author = "I.C.M. Rosseti",
        title = "Heur\'{\i}sticas para o problema de s\'{\i}ntese de redes
	         a 2-caminhos",
	school = "Department of Computer Science, Catholic University of
                  Rio de Janeiro",
        address = "Rio de Janeiro, Brazil",
	month = "July",
	annote = {Given a connected undirected graph $G=(V,E)$, 
		a nonnegative weight function defined on $V$, and 
		a set $D$ of origin-destination pair, the $2$-path 
		network design problem ($2$PNDP) consists in finding 
		a minimum weighted subset of edges containing a $2$-path 
		between the endpoints of every origin-destination in $D$, 
		where a $2$-path between the pair $(s,t)\in D$ 
		is a sequence of at most two edges connecting $s$ to $t$. 
		For solving $2$PNDP, sequential and parallel heuristics 
		are proposed, included variants and combinations of 
		{GRASP}. In Portuguese.},
        year = "2003"
}

@inproceedings{SriRamKumAraNaq00a,
	author = "A. Srinivasan and K.G. Ramakrishnan and K. Kumaram and
		  M. Aravamudam and S. Naqvi",
	title = "Optimal design of signaling networks for {Internet} telephony",
        booktitle = "IEEE INFOCOM 2000",
        city= "Tel-Aviv, Israel",
        month = "March",
        year = "2000",
	annote = 	"This paper presents an approach for efficient
                	design of a signaling network for a network of
                	software switches supporting Internet telephony.
                	Optimal load balancing for given demand forecast
                	is formulated as a quadratic assignment problem,
                	which is solved with a {GRASP}."
}

@techreport{VieGon01a,
        author = {C.E.C. Vieira and P.R.L. Gondim},
	title = {Uma Nova Estrat\'egia para Aplica\c{c}\~ao do {GRASP}
                  ao problema de aloca\c{c}\~ao de canal},
        number = {070/DE9/01},
	institution = {Departamento de Engenharia de Sistemas,
                       Instituto Militar de Engenharia},
        address = {Rio de Janeiro, Brazil},
	annote = {A special location problem arising in telecommunications 
		is addressed in this paper. 
		The authors propose a heuristic that uses the {GRASP} 
		randomized construction idea and 
		the chosen greedy function measures the difficulty 
		in assigning a frequency.},
	year = {2001}
}

@inproceedings{XuChiu96a,
	author = "J. Xu and S. Chiu",
	title = "Solving a real-world field technician scheduling problem",
	booktitle = "Proceedings of the International Conference on
               Management Science and the Economic Development of China",
        city = "Hong-Kong",
	month = "July",
	year = "1996",
	pages = "240--248",
	annote =	"The objective of the field technician scheduling 
			problem is to assign a set of jobs at different 
			locations with time windows to technicians
                	with different job skills. The greedy choice of the
                	proposed {GRASP} is to select jobs with the highest 
			unit weight.
                	The local search implements four different moves, among
                	them the $2$-exchange and a swap that exchanges an 
			assigned job with another job unassigned under the 
			candidate schedule."
}

@Article{XuChi01a,
	author = "J. Xu and S. Chiu",
	title = "Effective heuristic procedure for a field technician
                 scheduling problem",
	journal = "Journal of Heuristics",
	volume = "7",
	pages = "495--509",
	year = "2001",
        annote   =     "The objective of the field technician scheduling problem
                	is to assign a set of jobs
                	at different locations with time windows to technicians
                	with different job skills. Several heuristics
                	are designed and tested for solving the problem: a pure
                	greedy heuristic, a {GRASP}, and a local search algorithm.
                	The greedy choice of the {GRASP} proposed is to select
                	jobs with the highest unit weight.
                	The local search implements four different moves, among
                	them the $2$-exchange and a swap that exchanges an assigned
                	job with another job unassigned under the candidate schedule.",
}