NivaarExam PrepOfficial exam papers ↗

22-Mec-B4 Integrated Manufacturing Systems · May 2014

Question 4 of 8: Lot-Sizing a Requirements Schedule — EOQ, Periodic Reorder and Part-Period Balancing

Nivaar worked solution (AI-drafted; not reviewed by a licensed engineer)

Notes on this paper

Paper format. National Exams, May 2014 — 07-Mec-B4 Integrated Manufacturing Systems, 3 hours, OPEN BOOK, any non-communicating calculator permitted. Eight questions are printed; any five constitute a complete paper and each question is of equal value (20 marks). Only the first five answers appearing in the answer book are marked. All eight are solved here.

Reference texts. R. Chase and F. R. Jacobs, Operations and Supply Chain Management, 16th ed. (forecasting, work measurement, break-even, process control); S. Nahmias and T. Olsen, Production and Operations Analysis, 7th ed. (lot sizing, inventory control); E. S. Buffa and R. K. Sarin, Modern Production / Operations Management, 8th ed. (the requirements-schedule lot-size comparison of Question 4); D. C. Montgomery, Introduction to Statistical Quality Control, 8th ed. (Shewhart charts and capability); M. P. Groover, Automation, Production Systems, and Computer-Integrated Manufacturing, 5th ed. (materials handling, group technology coding, CAPP and CAD); B. W. Niebel and A. Freivalds, Methods, Standards, and Work Design, 13th ed. (time study, allowances, wage-incentive plans).

Question 4: Lot-Sizing a Requirements Schedule — EOQ, Periodic Reorder and Part-Period Balancing (20 marks)

Question text not reproduced: the examination questions are © Engineers and Geoscientists BC. Open the official past paper (linked at the top of this page) to read the question, then follow the worked solution below.

Given. A twelve-week discrete requirements schedule for a dependent-demand component, with a set-up (ordering) charge per lot and a carrying charge per unit held at the end of each week.

Given data
ItemSymbolValue
Average requirement (as printed)$\bar{R}$116.7 units per week
Preparation / set-up cost per lot$c_{p}$$400 per lot
Holding cost$c_{H}$$4 per unit per week
Planning horizon—12 weeks
Weekly requirements$R_{t}$25, 30, 75, 125, 200, 325, 400, 100, 0, 100, 0, 100 units
Sum of the tabulated requirements$\sum R_{t}$1,480 units (the summary line prints 1,390 — see the check note)

Find. The week-by-week inventory record (lot receipts and ending balances) and the total incremental cost, meaning set-up cost plus carrying cost over the twelve weeks, under each of the three lot-size policies, and hence which policy is cheapest for this schedule.

Twelve-week requirements schedule, motor drive unit090180270360450Requirement, units25130275312542005325640071008091001001110012triangles mark the weeks in which part-period balancing places a lot
The schedule is strongly lumpy - it rises from 25 units in week 1 to 400 in week 7 and then collapses - which is precisely the condition under which a fixed economic lot size performs badly and a discrete lot-sizing rule performs well.

Approach. Compute the fixed economic lot size and the economic order interval from the classical square-root formulas, run each policy week by week against the tabulated requirements to obtain its inventory record, and cost each record as (number of lots) times the set-up charge plus (sum of the weekly ending balances) times the carrying charge.

  1. Part (a) — compute the economic lot size. Balancing the set-up cost against the carrying cost of the average cycle stock gives the classical square-root lot size $$Q_{o}=\sqrt{\frac{2\,\bar{R}\,c_{p}}{c_{H}}}=\sqrt{\frac{2(116.7)(400)}{4}}=\sqrt{23{,}340}=152.8\;\Rightarrow\;\boxed{Q_{o}=153\ \text{units per lot}}$$ The lot is rounded to a whole unit; the total-cost curve is very flat near its minimum, so nothing is lost by the rounding.
  2. Run the fixed lot against the schedule. The rule is mechanical: enter each week with the balance carried forward, and if it cannot cover that week's requirement, receive as many whole lots of 153 units as are needed. Week 1 opens empty, so one lot arrives and 128 units are left over; weeks 2 and 3 are met from stock; week 6 needs 325 against an opening balance of 4, so three lots (459 units) arrive; week 7 needs two more.
    Inventory record, policy (a): fixed lot of 153 units
    Week123456789101112Total
    Requirement253075125200325400100010001001480
    Lot received153——153153459306153—153——1530
    Ending balance1289823514138449797150150501030
    Carrying cost$512$392$92$204$16$552$176$388$388$600$600$200$4,120
  3. Cost the fixed-lot record. Ten lots are released, and the ending balances sum to 1,030 unit-weeks: $$TC_{a}=10(400)+1{,}030(4)=4{,}000+4{,}120=\boxed{\$8{,}120}$$ The fixed lot is a poor fit here. It was derived from an average requirement, but the actual schedule is lumpy, so the policy leaves 128 units sitting through weeks in which almost nothing is used, and still needs three lots in a single week when demand spikes.
  4. Part (b) — convert the lot size into an order interval. The economic periodic reorder model orders at a fixed interval rather than in a fixed quantity, and the economic order interval is the lot size expressed in weeks of average demand: $$EOI=\frac{Q_{o}}{\bar{R}}=\sqrt{\frac{2\,c_{p}}{c_{H}\,\bar{R}}}=\sqrt{\frac{2(400)}{4(116.7)}}=1.31\ \text{weeks}\;\Rightarrow\;\boxed{T=1\ \text{week}}$$ Since orders can only be placed at the weekly buckets of the schedule, the interval is rounded to the nearest whole week.
  5. Run the periodic policy. With a one-week interval the rule degenerates into lot-for-lot ordering: each week receives exactly its own requirement and nothing is carried, except that the two weeks with no requirement need no order at all.
    Inventory record, policy (b): reorder every week, order the week's requirement
    Week123456789101112Total
    Requirement253075125200325400100010001001480
    Lot received253075125200325400100—100—1001480
    Ending balance0000000000000
    Carrying cost————————————$0
    Ten lots are released and nothing is ever held, so $$TC_{b}=10(400)+0(4)=\boxed{\$4{,}000}$$ For comparison, rounding the interval up to two weeks instead would give six lots and 780 unit-weeks of stock, $2,400 plus $3,120, a total of $5,520 — so the one-week rounding is the correct one.
  6. Part (c) — find the economic part-period. Part-period total cost balancing extends each lot forward in time until the cost of carrying the added requirement is about to exceed the set-up cost that would be saved. The balance point is the economic part-period $$EPP=\frac{c_{p}}{c_{H}}=\frac{400}{4}=\boxed{100\ \text{part-periods}}$$ meaning that carrying 100 units for one week, or one unit for 100 weeks, costs exactly as much as one additional set-up.
  7. Accumulate part-periods lot by lot. Starting in week 1, adding week 2 to the lot costs $30\times1=30$ part-periods; adding week 3 as well would cost $30+75\times2=180$, which overshoots 100 by more than 30 undershoots it, so the first lot covers weeks 1 and 2 only, 55 units. The same test applied from week 3 gives $125\times1=125$ part-periods, nearer to 100 than 0 is, so the second lot covers weeks 3 and 4, 200 units. Continuing in this way places lots in weeks 1, 3, 5, 6, 7, 10 and 12.
    Inventory record, policy (c): part-period balancing, EPP = 100
    Week123456789101112Total
    Requirement253075125200325400100010001001480
    Lot received55—200—200325500——100—1001480
    Ending balance30012500010000000255
    Carrying cost$120—$500———$400—————$1,020
  8. Cost the part-period record and compare the three policies. Seven lots are released and the ending balances sum to 255 unit-weeks: $$TC_{c}=7(400)+255(4)=2{,}800+1{,}020=\boxed{\$3{,}820}$$ which is the cheapest of the three. The week-10 decision is a genuine tie on the part-period test — extending that lot to cover week 12 as well would accumulate 200 part-periods, exactly as far above 100 as 0 is below it — and the tie is broken by costing both: covering weeks 10 to 12 in one lot saves one set-up ($400) but adds 200 unit-weeks of stock ($800), a net penalty of $400. Placing the separate week-12 lot is therefore correct.
Question 4 — total incremental cost by policy
PolicyLot ruleLotsSet-up costCarrying costTotal incremental cost
(a) Economic lot sizefixed lot $Q_{o}=153$ units10$4,000$4,120$8,120
(b) Economic periodic reorderorder every $T=1$ week10$4,000$0$4,000
(b) alternative, $T=2$ weeksorder every 2 weeks6$2,400$3,120$5,520
(c) Part-period balancing$EPP=100$ part-periods7$2,800$1,020$3,820

Check: the printed twelve-week total does not match the printed schedule. The requirements listed in Table 1 sum to 1,480 units, not the 1,390 stated beneath it, and neither figure is consistent with the given average of 116.7 units per week (which implies about 1,400 over twelve weeks). The inventory record can only be built from the week-by-week requirements, so those are used as printed; the given $\bar{R}=116.7$ is used as printed in the two square-root formulas, as the question directs. Using the tabulated average of 123.3 units per week instead would give $Q_{o}=157$ units and $EOI=1.27$ weeks — the same rounded interval, the same policy ranking and the same conclusion, so the discrepancy does not affect any decision in the answer.