Scribd
Upload
24 views29 pages

Day & Shift

Workforce scheduling involves assigning employees to shifts while considering various constraints and preferences. The document discusses using an algorithm called PEST to schedule nurses at an intensive care unit in a Finnish hospital. PEST stands for Population, Election, Annealing, Shuffling, and Tau and is used to explore promising solutions while avoiding getting stuck in local optima. The goal is to minimize understaffing and overstaffing while respecting constraints such as ensuring sufficient competencies per shift and not assigning consecutive shifts over 9 days.

Uploaded by

Dharun Blazer
Copyright
© © All Rights Reserved
We take content rights seriously. If you suspect this is your content, claim it here.
Available Formats
Download as PPTX, PDF, TXT or read online on Scribd
24 views29 pages

Day & Shift

Workforce scheduling involves assigning employees to shifts while considering various constraints and preferences. The document discusses using an algorithm called PEST to schedule nurses at an intensive care unit in a Finnish hospital. PEST stands for Population, Election, Annealing, Shuffling, and Tau and is used to explore promising solutions while avoiding getting stuck in local optima. The goal is to minimize understaffing and overstaffing while respecting constraints such as ensuring sufficient competencies per shift and not assigning consecutive shifts over 9 days.

Uploaded by

Dharun Blazer
Copyright
© © All Rights Reserved
We take content rights seriously. If you suspect this is your content, claim it here.
Available Formats
Download as PPTX, PDF, TXT or read online on Scribd
You are on page 1/ 29

Workforce scheduling

Bin Biddappa K S 13MF31


Vignesh 13MF36
WORKFORCE is all aou! assigning
e"plo#ees !o shif!s$ de!er"ining %orking
da#s and res! da#s or cons!ruc!ing
&e'ile shif!s and !heir s!ar!ing !i"es
Workforce scheduling consis!s roughl# of
e(er#!hing fro" de!er"ining !he needs of
!he cus!o"ers !o de!er"ining !he e'ac!
schedule of each e"plo#ee)

*a#s+o, scheduling deals %i!h !he


assign"en! of res! da#s e!%een
%orking da#s o(er a gi(en planning
hori-on)

*a#s+o, scheduling also includes !he


assign"en! of (aca!ions and special
da#s$ such as union s!e%ard du!ies
and !raining sessions)
Nurse Rostering in an Intensive-Care Unit
in a Finnish Hospital

.his prole" includes /(e charac!eris!ics !ha! are


no! al%a#s presen! in !he nurse ros!ering case
repor!ed in !he acade"ic li!era!ure0

.he nu"er of nurses is o(er 111

.he nurses are grouped in four ca!egories ased on


!heir !o!al %orking hours %i!hin !he planning hori-on)

So"e shif!s las! "ore !han 12 hours and ac!uall#


include !%o consecu!i(e shif!s

So"e nurses should al%a#s %ork on !he sa"e shif!s

.he nurses3 %ishes for da#s+o, and shif!s co(er as


"uch as 415 of !heir !o!al %ork)

.he "os! i"por!an! goal is !o


"ini"i-e unders!a6ng and
o(ers!a6ng) 7o%+8uali!# ros!ers can
lead ei!her !o an undersuppl# of
nurses %i!h a need !o hire par!+!i"e
nurses or an o(ersuppl# of nurses
%i!h !oo "uch idle !i"e$ i"plica!ing
a loss of e6cienc#
Cons!rain!s in *a#+O,
Coverage requirement

9C1: ;n e"plo#ee canno! e assigned


!o o(erlapping shif!s

9C<: ; "ini"u" nu"er of e"plo#ees


%i!h par!icular co"pe!ences "us! e
guaran!eed for each shif!

9C2: ; alanced nu"er of surplus


e"plo#ees "us! e guaran!eed in each
%orking da#
Regulatory requirements

9R1: .he re8uired nu"er of %orking da#s$


%orking hours and da#s+o, %i!hin a
!i"efra"e "us! e respec!ed

9R<: .he re8uired nu"er of holida#s


%i!hin a !i"efra"e "us! e respec!ed

9R3: .he re8uired nu"er of free


%eekends 9o!h Sa!urda# and Sunda# free:
%i!hin a !i"efra"e "us! e respec!ed

9R4: .he "ini"u" !i"e gap of res!


!i"e e!%een !%o shif!s "us! e
respec!ed

9R6: .he nu"er of special shif!s


9such as union s!e%ard du!ies and
!raining sessions: for par!icular
e"plo#ees %i!hin a !i"efra"e "us!
e respec!ed

9R=: E"plo#ees canno! %ork


consecu!i(el# for "ore !han w days
Operational requirements

9O1: ;n e"plo#ee can onl# e assigned


!o a shif! he>she has co"pe!ence for

9O<: ;! leas! g working days must be


assigned between two separate days-
of breaks

9O4: ;n e"plo#ee assigned !o a shif!


!#pe t1 must not be assigned to a shift
type t2 on the follo%ing da#
Operational preferences

9E1: Single da#s+o, should e a(oided

9E<: Single %orking da#s should e


a(oided

9E3: .he "a'i"u" leng!h of


consecu!i(e da#s+o, is d

9E2: ; alanced assign"en! of single


da#s+o, and single %orking da#s "us!
e guaran!eed e!%een !he e"plo#ees

9E4: ; alanced assign"en! of


di,eren! shif! !#pes "us! e
guaran!eed e!%een !he e"plo#ees

9E=: ; alanced assign"en! of


%eekda#s "us! e guaran!eed
e!%een e"plo#ees

9E?: ;ssign or a(oid a gi(en shif!


!#pe efore or af!er a free period
9da#s+o,$ (aca!ion
Personal preferences

9@1: ;ssign or a(oid assigning gi(en


e"plo#ees !o !he sa"e shif!s

9@<: ;ssign a re8ues!ed da#+on or


a(oid a re8ues!ed da#+o,

9@3: ;ssign a re8ues!ed shif! or a(oid


an un%an!ed shif!
Shift generation

Shif! genera!ion !ransfor"s !he


de!er"ined %orkload in!o shif!s) .his
includes deciding reak !i"es %hen
applicale)

Shif! genera!ion is essen!ial


especiall# in cases %here !he
%orkload is no! s!a!ic)

; asic shif! genera!ion prole"


includes a (ariale nu"er of
ac!i(i!ies for each !ask in each
!i"eslo!
Cons!rain!s in SA
Structural constraints

9SAS1: Bo shif! should con!ain


!i"eslo!s %i!h "ul!iple !#pes of
ac!i(i!ies 9i)e) di,eren! !asks or o!h
reaks and !asks:)

9SAS<: Bo shif! should con!ain gaps,


i.e. timeslots with no activities.
Coverage constraints:

9SAC1: .he nu"er of e"plo#ees a!


each !i"eslo! o(er !he planning
hori-on "us! e e'ac!l# for each
s!ric!l# !i"e+dependen! !ask)

9SAC<: .he su" of !he e'cesses of


e"plo#ees a! each !i"eslo! o(er !he
planning hori-on "us! e "ini"i-ed
for each s!ric!l# !i"e+dependen! !ask)

9SAC3: .he su" of !he shor!ages of


e"plo#ees a! each !i"eslo! o(er !he
planning hori-on "us! e "ini"i-ed
for each s!ric!l# !i"e+dependen! !ask)

9SAC2: .he !o!al %orkload for each


!ask "us! e carried ou! efore a
gi(en !i"eslo!)

9SAC4: Bo %orking !i"e should e


%as!ed)
Volume constraints

9SAV1: .he nu"er of shif!s$ i)e) !he


nu"er of e"plo#ees a! %ork$ "us!
e "ini"i-ed)

9SAV<: Shif!s of e'ac!l# k timeslots in


length must be maximied.

9SAV3: .he leng!hs of !he shif!s "us!


"a!ch !he e"plo#ees3 a(ailale
hours)

9SAV2: .he co"pe!ences necessar#


!o carr# ou! !he shif!s "us! "a!ch
!he a(ailale %orkforce)
Placement constraints

9SA@1: Each shif! "us! con!ain a


gi(en nu"er of reaks of cer!ain
leng!hs$ depending on !he leng!h of
!he shif!)

.he planning hori-on is si' %eeks) .he


!o!al %orking hours for each full+!i"e
nurse are <<C hours and 31 "inu!es 9R1:)

.he %orking hours can also e 1?1h$


112h 24"in or 1?1h 31"in if a nurse is
on par!+!i"e pension or has s"all
children) .he holida#s 9R<: and special
shif!s 9R6: are included in !he %orking
hours)
The PEST lgorithm
.he acron#" @E;S. s!e"s fro" !he
"e!hods used as

@opula!ion$

EDec!ion$

;nnealing$

ShuEing

.au)
ODec!i(e of @E;S.
1: .o e'plore pro"ising areas in !he
search space !ha! is$ !o -oo"+in !o
/nd local op!i"u" solu!ions !o a
su6cien! e'!en! %hile a! !he sa"e
!i"e
<: ;(oiding s!a#ing s!uck in !hese
areas for !oo long and
3: Escaping fro" !hese local op!i"a in
a s#s!e"a!ic %a#
Conclusion

;s !he hard cons!rain!s s!a!ed$ !he "os!


i"por!an! goal %as !o /nd a solu!ion !ha!
has no o(erlapping shif!s and guaran!ees
a su6cien! nu"er of co"pe!ences for
each shif!$ and %here e"plo#ees do no!
%ork consecu!i(el# for "ore !han nine
da#s$ ha(e su6cien! res! !i"e e!%een
shif!s and are no! assigned !o a foridden
shif! efore>af!er a nigh! shif!)
Reference

Bico R) M) K#ngFs$ Ki""o G) Bur"i$ and Gari R)


K#ngFs H.he Workforce Scheduling @rocess
Ising !he @E;S. ;lgori!h" Hin JMECS <113$
March 13 + 14$ <113$ Kong Kong)

Bico K#ngFs1$ Ki""o Bur"i1$ E#DLlfur Jngi


Msgeirsson<$ Gari K#ngFs1N Ising !he @E;S.
;lgori!h" !o Ros!er Burses in an Jn!ensi(e+Care
Ini! in a Finnish Kospi!alN in @rac!ice and
.heor# of ;u!o"a!ed .i"e!aling 9@;.;. <11<:$
<C+31 ;ugus! <11<$ Son$ Bor%a#
.hank Oou

You might also like