Systems Thinking and the Theory of Constraints

Document Sample
Systems Thinking and the Theory of Constraints Powered By Docstoc
					Timo Hildebrandt
Paul Sullivan
Alice Wang
Outline of Topics
 Timo Hildebrandt
     Theory of Constraints (TOC)

   Alice Xin Wang
     Activity Based Costing (ABC)

   Paul Sullivan
     Theory of Constraints the Throughput World
   Eli Goldratt, an Israeli physicist , created the Theory of
    Constraints when a friend asked for help scheduling his
    chicken coop business (i.e. a chicken house).

   He developed a scheduling algorithm that realistically
    considered the bottlenecks in the system.

   In 1979 he created a computer software which using
    mathematical programming and simulation , allowed for
    more complex scheduling of manufacturing systems.

   Optimized Production Technology (OPT)
   The idea behind the software is that in any process,
    there is one set or at most a few bottlenecks.

   The output of a process is constrained by the

   You need to focus on maximizing the productive use
    of the bottleneck and schedule it accordingly, and
    the non-bottleneck resources can be scheduled
   “Just like the links of a chain work together to pull or lift objects, the
    processes within the enterprise work together to generate profit for the
    shareholders. However, the ability of the chain to pull or lift objects is
    limited because the chain is only as strong as its weakest link. TOC
    maintains that:
     The goal of an enterprise is to make more money, now and in the
     Every system is interdependent process is subject to at least one
       constraint that limits system performance. That is, the money-
       making potential of the enterprise is limited by the enterprise’s
       weakest link or constraint.”
   Time lost at a bottleneck resource results in a
    loss of productivity for the whole enterprise
    (entire supply chain). Time saved a non-
    bottleneck resources is a mirage.
Local Optimization:
 Address the complex problem of optimizing system
  performance by breaking down the system into smaller,
  more manageable pieces.
 Units were treated as either cost centers or profit centers
  and assigned targets.
 The logic was that if every unit improved then, ipso facto, the
  entire enterprise would improve. (Lean Supply Chain
  Principle 1)
 However it is not enough to promote isolated efforts that
  focus on improving specific functions.
Global Optimization:
 Install measures that are truly global, i.e., measures that
    encourage actions consistent with the overall goals of the
   These measures are based on the a throughput world
    perspective, an approach that focuses on meeting the goals
    through a growth strategy rather than through cutting costs.
   There are two groups of techniques:
     A Five-Step Focusing Process
     Thinking Process
 Step 1: Identify the System’s Constraint(s)
 Step 2: Decide how to Exploit the
          System’s Constraints
 Step 3: Subordinate Everything Else to
          that Decision
 Step 4: Elevate the System’s Constraints
 Step 5: If a Constraint Was Broken in
          Previous Steps, Go to Step 1
 Physical Constraints
      Physical, tangible; easy to recognize as constraint. Machine
       capacity, material availability, space availability, etc.
 Market Constraints
      Demand for company’s products and services is less than
       capacity of organization, or not in desired proportion.
 Policy Constraints
      Not physical in nature. Includes entire system of measures and
       methods and even mindset that governs the strategic and
       tactical decisions of the company.
   Mindset Constraints
     A constraint if thought process or culture of the organization
      blocks design & implementation of measures & methods
      required to achieve goals
   Measures Constraints
     A constraint if the measurement system drive behaviors that are
      incongruous with organizational goals
   Methods Constraints
     A constraint when procedures and techniques used result in
      actions incompatible with goals
   Littlefield Technologies
     Time: Limited duration to make changes and
      experiment with system design
     Capital: Cash to procure machines and raw
     Resources: Under utilization and over utilization
     Strategy: Decisive actions taken as early as
      possible to benefit profit maximization
   Focuses on minimizing costs and improving
    operational efficiencies.
     Roots in the 20th Century when U.S. Industry grew
     U.S. Industry began to borrow money to pay for resources, bankers
      and lenders needed to know how these resources would perform.
     Determining product costs is important in deciding whether the
      product was profitable or not.
   Products costs have three components:
     Direct labor
     Direct Material
     Overhead Costs
   How do we allocate overhead costs to
    product costs so that product costs are

   Cost Accounting
     Standard Costing
     Activity Based Costing
 A standard set of guidelines and operating
  procedures used to decide whether or not to
  approve purchases of new equipment or scheduling
  of equipment.
 The cost world directs the manager to think locally,
  but the throughput world forces the manager to
  think globally.

   Lets look at an example…
   Assumptions:
     Only 1 product – (Black Vinyl Steering Wheels)
     Raw materials cost $10/unit – (Blanks)
     Uses two resources to machine and fabricate the product –
      (Resource A & Resource B)
     6 minutes processing at Resource A
     9 minutes processing at Resource B
     Run’s 2 shifts 10 hours long, 25 days each month
     Employs 4 direct laborers across the 2 shifts
   Each operator works for 25 x 10 = 250 hours per month
   Direct laborers are paid $2,500 (250 hrs * $10/hr)
   The total wages paid per year are ($2500 * 4) * 12 or $120,000 ($10,000 *
   The manufacturing overhead costs were $225,000
   Selling, General and Administrative costs were $200,000
   The standard cost for the product is the sum
    of the standard direct labor cost (the
    absorbed labor cost), plus the unit raw
    material cost, plus a unit overhead cost that
    accounts for the manufacturing overhead.

   Direct labor is a Fixed Cost (FC) because it
    does not rise or fall with production like the
    raw material cost (Variable Cost V.C.).
   Direct labor costs at Resource A
     Hourly rate $10 * time at resource 0.10 hours
      ▪ $10 * 0.1 = $1.00 per unit

   Direct Labor costs at Resource B
     Hourly rate $10 * time at resource .15 hours
      ▪ $10 * 0.15 = $1.50 per unit

   Total Direct Labor Costs = $2.50
   Average Demand = 40,000 steering wheels

   2003 Demand = 30,000 steering wheels

   Direct Labor Costs =
     30,000 units * $2.50/unit = $75,000

   Actual Labor Costs =
     $120,000
   The difference between actual wages paid
    and the overhead labor attached to the units

   $120,000 - $75,000 = $45,000

   That means $45,000 of direct labor was not
    accounted for in product costs
   Typically expressed in terms of direct labor
    hours or direct labor costs
     $225,000/$75,000 = $3.00 per direct labor dollar
   Since the direct labor cost is $2.50 the
    overhead per unit is
     $3.00 * 2.50 = $7.50 per unit
   The finished product cost is
    $10 (direct material) + $2.50 (direct labor) + $7.50 (overhead) = $20.00
Cost Category                                     Cost per Unit
Raw Material                                      $10.00

Direct Labor Cost at Resource A                   $1.00

Manufacturing Overhead allocation at Resource A   $3.00

Direct Labor Cost at Resource B                   $1.50

Manufacturing Overhead allocation at Resource B   $4.50

Total Cost                                        $20.00

 The raw material in front of resource A is $10.00/unit; the value of a WIP present
 between Resource A and Resource B is $14.00 ($10 + $1 + $3); hence the finished
 goods inventory is valued at $14 + $6 ($1.50 + $4.50) = $20!
   Captures the effort expended on the product
    in dollar terms
     This is an approximate method because it
     assumes that every manufacturing-related
     expense, including depreciation of equipment and
     facilities, indirect labor, and utilities is allocated to
     individual resources. (This doesn’t include SG&A
   Focuses on increasing value of the product
   Incentive for shop supervisor to push
    products through the resources as quickly as
   Contrary to a Kanban system because
    downstream activities may not need these
   Leads to End-of-the-Month syndrome
   Results in even higher levels of inventory
   “Under absorption costing, building inventories
    tends to reduce the apparent average COGS.
    When production exceeds sales, fixed costs are
    spread across more units, and some of the fixed
    costs are reported on the balance sheet as part
    of additional inventories rather than on the
    income statement as part of COGS. Under
    standard cost variance reporting, a work center
    with a fixed labor force can improve its efficiency
    measure only by producing more output.”
    Why ABC?
    - ABC provides an opportunity to obtain a
     better approximation of the true product cost

   Basic Principle of the ABC system is
    “Activities consume costs and product
    consume activities”
   ABC focuses on cost drivers that can guide
    the allocations with greater accuracy.
   Cost drivers are allocated into one of four
    - Unit-lever activities: activities performed for each product. (Example:
    direct labor, direct material cost)
    - Batch-level activities: activities that are performed once for each batch
    of products. (Example: machine setup costs, material handling costs)
    - Product-level activities: activities support the production of a product
    type or model. (Example: engineering support costs, depreciation cots of
     - Facility-level: include the costs of operating the Accounting, HR,
    General Administration, Sales, and Plant Maintenance functions.
Cost Category     Standard Costing        ACTIVITY-BASED
                  Vinyl       Wood-Trim   Vinyl     Wood-Trim
Raw Material      $ 10.00     $ 10.00     $10.00    $10.00
Direct Labor      $ 2.50      $ 2.50      $ 2.50    $ 2.50
Depreciation      $ 0.40      $ 0.40      $ 0.40    $ 0.40

Mfg. O/H (QC      $ 2.10      $ 2.10      $ 2.10    $ 2.10
Mfg. O/H (Plant   $5.00       $ 5.00      $ 1.50    $ 22.50
Product Cost      $ 20.00     $ 20.00     $ 16.50   $ 37.50
   The Standard cost accounting system allocates all overhead
    costs to the products using a broad measure such as labor

   A more equitable allocation of overhead costs – ABC system
    can help approximate the product cost

   The problems of ABC systems

   ABC system can become difficult and expensive to maintain
   Two Premises of TOC:
    - The goal of a business is to make more
    money not and in the future
    - A system’s constrain determine its output

   Constraint : anything that inhibits a system’s
    performance toward its goals
   TOC encourages systems thinking
   TOC adopts the view that managers should
    focus on a system’s constraints as a basis for
    making the right decisions
   TOC prescribes three performance measures
     - Throughput
     - Inventory
     - Operating expense
   Throughput (T): the rate at which the system
    generates money through sales

   Inventory (I): All the money invested in
    purchasing things the system intends to sell

   Operating expense (OE): All the money the
    system spends, turning inventory (I) into
    throughput (T)
 Asset base as of January            Activity in 2003
   1, 2003                     RM Purchased in 2003       40,000

Fixed Assets        $120,000   Number of Units Sold       30,000
                               in 2003
Raw Material          25,000   Unit Selling Price           $30
Inventory (Units)
WIP Inventory         15,000   Direct Labor Wages       $120,000
(Units)                        Plant Manager’s          $150,000
Finished Goods        20,000   Salary
Inventory (Units)              QC Inspector’s Salary     $63,000

                               SG&A Expenses            $200,000

                               Capital Assets            $25,000
                               Acquired in 2003
                               Depreciation in 2003      $12,000
 Asset base as of December 31, 2003
Fixed Assets                   $133,000

RW Inventory (Units)             35,000

WIP Inventory (Units)            15,000

Finished Goods Inventory         20,000
I (start of 2003) = $720,000
T ( in 2003)      = $600,000
OE (in 2003)      = $545,000
I (end of 2003) = $833,000

Profit T –OE = $55,000
   Bottom Line Measurements


     (Absolute)    RETURN ON
                                 CASH FLOW

   Quite different from traditional cost
    accounting measures
     Throughput (T) = The rate at which the system
      generates money through sales
     Inventory (I) = All the money invested in
      purchasing things needed by the system to sell its
     Operating Expense (OE) = All the money the
      system spends, turning inventory into
 Throughput (T) = Sales Revenue - Variable
     Cost (Materials)
 Operating Expenses (OE) = Fixed Costs
     (Direct Labor + Overhead)
These can be equated with traditional
     financial measures
   Net Profit = Throughput – Operating Expense =       T – OE
   Return on Investment = Throughput – Operating Expense =
                         T – OE                Inventory
 Inventory Turns = Throughput = T
                         Inventory = I
   Productivity =       Throughput = T
                 Operating Expense OE
   TOC approach less than Traditional especially in Inventory
    turns b/c denominator in TOC includes all capital assets
Priority   Cost World   Throughput World
 First:         OE             T
Second:         T               I
Third:          I

                             OE  distant
                             3rd Priority
   Uses two resources A and B for production
   Resource A = Six minutes p/unit
   Resource B = Nine minutes p/unit
   Each resource operated by a direct labor employee
   Worked two 10 hr shifts with four direct laborers
   In 2003, produced 30,000 units w/ selling price of $30 p/unit
   Time Required at Resource B = 30,000 x (9/60) = 4,500 hrs/yr
    (This is the Capacity Constraint)
   Resource B is available 20 hrs/day and 25 days/month or
    500 x 12 = 6,000 hrs/yr
   In 2003, National Pontoons operated at 75% capacity 
    4,500 / 6,000 hrs/yr
Cost and Profit Data for National Pontoons without Leveraging

  ITEM                                      DATA FOR 2003
  Sales Revenue                             $900,000
  Raw Material Cost                         $300,000
  Direct Labor Cost                         $120,000
  Overhead Costs                            $425,000
  Total Expenses                            $845,000
  Net Profit                                $55,000
   Next, Focus on Leveraging Power of OE to
    become more profitable in 2004
   Only operating at 75%, reduces working
    hours by 25% in 2004 to 15 hours per day
   Adjusts wages paid to direct labor by 25%
   Labor drops from $120,000 to $90,000 
    $120,000 x 0.75
   The $30,000 cost savings goes directly to
    bottom line
Leveraging Power of OE for National Pontoons
  ITEM                    DATA FOR         LEVERAGING
                          2003             POWER OF OE
  Sales Revenue           $900,000         $900,000
  Raw Material Cost       $300,000         $300,000
  Direct Labor Cost       $120,000         $90,000
  Overhead Costs          $425,000         $425,000
  Total Expenses          $845,000         $815,000
  Net Profit              $55,000          $85,000
   Next, Focus on using its installed capacity to
    improve T to become more profitable in 2004
   Starts marketing campaign to sell more units
    which drops price 5% and boosts its sales to
    40,000 units in 2004
   Raw Materials increased by 33.33% to
   Selling price  0.95 x $30 = $28.50 p/unit
   Results in full capacity utilization at Resource
    B  40,000 x (9/60) = 6,000 hrs/yr
Leveraging Power of T for National Pontoons
                 2003           POWER OF OE   POWER OF I
Sales Revenue    $900,000 $900,000            $1,140,000
Raw Material     $300,000     $300,000        $400,000
Direct Labor     $120,000     $90,000         $120,000
Overhead Costs   $425,000     $425,000        $435,000
Total Expenses   $845,000     $815,000        $955,000
Net Profit       $55,000      $85,000         $185,000
   Example suggests that the leveraging power
    of Throughput (T) exceeds the leveraging
    power of operating expense (OE)
     For most manufacturing enterprises, the material
      cost is a large percentage of the total product cost
     Enterprises that focus on reducing OE can only do
      so by trimming labor costs because it is harder to
      get rid of fixed assets and infrastructure
   A chain can demonstrate how the
    cost world focus sacrifices long term
    throughput increases for
    short-term gains
   Instead of strengthening the weakest link
    (improving T), we focus on improving
    efficiency at the current level of performance
    (improving OE)
   Let’s take a look at an example
 Assume a chain of 10 links
 Each link = 100 lbs of carrying capacity except for one
  link has only 50 lbs of carrying capacity
 Management unhappy with cost of maintaining the
  nine strong links so it sells the nine heavy links
 It replaces each link with a carrying capacity of 50
  lbs which makes a truly efficient chain since every link
  is capable of carrying exactly same load
 What is the problem?
   The problem is the enterprise is locked into the
    current performance level
   It now has ten links, any one of which can break
   In the future, if improved performance is desired, it
    will have to work all ten links in the chain
   The same problem occurs when enterprises eliminate
   If business picks up, it will be harder to recruit
    employees, why?
   Fearful of being fired in the next downsize
   Decisions should promote a growth strategy.
    While enterprises should attempt to
    simultaneously increase throughput,
    decrease inventory, and decrease operating
    expenses, the focus must be on improving
   What is this guy doing? 
 How do we leverage the growth model advocated by
  the throughput world ?
 What should we do when the enterprise finds it has
  excess capacity?
     In the National Pontoons example, the focus on T generated a
      much better return than did the cost world focus on OE.
     This is not the only approach one can pursue.
     Improving the bottom line is the best done with a logical
   The TOC presents a five-step focusing process for
    identifying the root cause of a problem and dealing with
    it effectively
   Step 1  Identify the System’s
   Step 2  Decide how to Exploit the
             System’s Constraint(s)
   Step 3  Subordinate Everything Else to
             that Decision
   Step 4  Elevate the System’s Constraints
   Step 5  If a Constraint Was Broken In a
             Previous Step, Go Back to Step 1
    The Five Step focusing process to improve
     system performance is based on two
     simple premises of TOC:
    1. The goal of the enterprise is to make more
       money, now and in the future
    2. The system’s constraints prevent it from
       making more money
    The TOC asserts that any enterprise has few
     real constraints, at most, a handful.
    The system’s constraints can be:
    1. Physical Constraints – Easiest to identify
    2. Market Constraints – Harder to Identify
    3. Policy Constraints –   Hardest to Identify

 The easiest to identify
 Can be machine capacity, staff availability,
  space availability, human capability, etc.
 Tangible
 Techniques for getting the most from
   capacity constraints:
     Eliminate periods of idle time
     Reduce setup time and run time per unit
     Improve quality control
     Purchase additional capacity
                    Identifying Physical Constraints:
                      A Typical WIP Inventory Profile:
Ave. WIP Inventory

                         R1      R2       R3      R4      R5   R6
   Harder to identify than physical constraints
   A market constraint exists if the demand for the
    enterprise’s products and services is less than the
    enterprise’s installed capacity or limits the
    bottomline performance
   Excess capacity is easily identified as a market
   Can arise when an enterprise produces and does
    not have enough capacity to satisfy all of the
   Not physical in nature.
   Includes entire system of measures and
    methods and even mindset that governs the
    strategic and tactical decisions of the
   Hardest to Identify
   In theory, there should be no policy
    constraints, but most constraints to system
    performance are policy constraints
   Methods Constraints
     A constraint when procedures and techniques used result in
      actions incompatible with goals
     Could be created by never producing a batch of units below an
   Measures Constraints
     A constraint if the measurement system drive behaviors that are
      incongruous with organizational goals
     Aggressively seeking quantity discounts would lead to increased
      raw material inventory
   Mindset Constraints
     A constraint if thought process or culture of the organization
      blocks design & implementation of measures & methods required
      to achieve goals
     Shop Supervisor has the attitude that all operators should be busy
      all of the time
 Exploiting the constraint means using the constraint as
  profitably as possible
 Until the constraints are overcome by other means, the
  enterprise should work them as profitably (effectively) as
  possible: the real meaning of the word exploit
     If constraint is physical resource, ensure the resource is never idle
     If the market is the constraint, exploit by ensuring not a single
      sale is lost as a result of our action or inaction
     Market constraint implies extra capacity, so we exploit this by
      guaranteeing 100% on time delivery to customer
 Need to manage every part of the profit-generating
 We do not want those non-bottleneck resources
  becoming bottlenecks because of our negligence
  because of focusing on constraints
 Deals with process of scheduling
     Work must be started and sequenced so the constraint can
      always work or work smarter
   Drum-buffer-rope (DBR), or pull-from-the-bottleneck
    model, is similar to kanban system
     Except the input process of DBR is linked to the rate or
      production of the constraint to utilize it as much as possible
 Tries to lift the restriction that is preventing the
  enterprise from making more money
 Elevating the constraint means identifying ways
  that the performance of the system can be
  improved, relative to its goals
 This step should only be performed after the exploit
  step, step 2
 If the constraint still exists, or another emerges,
  then it is time to execute the fourth step
   Can we stop with the fourth step?
   If we elevate the constraint, it will probably not
    remain a constraint
     Performance will not be dictated by another element that
      has become the weakest link
     To find this new weak link, we must revisit all the steps
      once again
   Goldratt adds this warning: “Do not allow inertia
    to cause a system’s constraints.”
   Step 5 is crucial because it prevents inertia from
    derailing continuous improvement process
   Think of a small production system for
    manufacturing two products, P and Q
   Goal is to maximize profit
   Weekly demand for P is 110 units & for Q is 60 units
   Four resources used to meet demand: A, B, C, and
   Let’s take a look at the process flow chart which
    includes the cost and time for each step
   Step 1 is executed to determine if a constraint
    prevents us from meeting the market demand
         A Production System Manufacturing Two Products, P and Q
                               $90 / unit                  $100 / unit
                    P:      110 units / week       Q:
                                                          60 units / week

                                      D                         D
   Purchased Part                  10 min.                    5 min.
      $5 / unit
                           C                      C                      B
                         10 min.                5 min.                 25 min.

                        A                         B                       A
                     15 min.                    10 min.                10 min.

                      RM1                       RM2                      RM3
                     $20 per                   $20 per                  $25 per
                       unit                     unit                     unit

Time available at each work center: 2,400 minutes per week
           Operating expenses per week: $6,000
Work       Loan on P      Load on Q     Total Time   Time
Center     (@110 units)   (@60 units)   Required     Available

A          1,650          600           2,250        2,400

B          1,100          2,100         3,200        2,400

C          1,650          300           1,950        2,400

D          1,100          300           1,400        2,400

   Do a Capacity analysis and we find Resource B is
 Have to Decide which product we want prioritized and produce it
 Product contribution is selling price less the price of materials used
  in the product
     Contribution of P = $90 - $5 - $20 - $20 = $45
     Contribution of Q = $100 - $20 - $25 = $55
   Cost World Analysis – work with product cost and product profits
     Q has higher profit margin ($55) than P ($45)
     With this approach, use the constraint to produce as many Q’s as possible and
      use remaining capacity to produce P
     Means producing 60 units of P (2100 minutes of Resource B), leaving 300
      minutes which is enough to produce 30 units of P
     Product Mix results in a net profit of $45(30) + $55(60) - $6,000 = -$1,350
   From a system’s perspective (throughput world), product costs do not
    have much weight for we are primarily in system’s profits
   To get maximum profits, we want to exploit the constraint
   Running Q through constraint, Resource B, requires 35 min/unit and Q’s
    profit margin is $55
     The Rate at which the constraint generates profit is $55/$35 =
      $1.57 p/min
   Running P through constraint, Resource B, requires 10 min/unit and P’s
    profit margin is $45
     The Rate at which the constraint generates profit is $45/10 =
      $4.50 p/min
   Shows we want to produce as many units of P as the market would bear
    and use remaining capacity on constraint for Q
   To produce 110 units of P needs 110 x 10 = 1,100 minutes of
    Resource B
   Since Time Available is 2,400, it leaves 2,400 – 1,100 = 1,290
    minutes available time on Resource B each week to
    produce Q
   With the available time, we can produce 1,290 / 35 which
    equals 36 units of Q
     Total Marginal Profit of Q  $45 x 110 + $55 x 36 = $6,930
 Since operating expense was $6,000, net profit = $930
 Cost World Approach  Net Loss of -$1,350
 Throughput Approach  Net Profit of $930
 Quite a difference!
 Relating to example, this step means keeping Resource B
  running at all times
 We can first have Resource B work on raw material RM2,
  during which Resource A would be processing 36 units of
  raw material RM3 each week to produce product Q
 If more than 36 units of A are produced, they would
  accumulate in front of Resource B
        The bottleneck has now been exploited
        Besides identifying Resource B as a bottleneck, we have
         found another: the Product Mix (water level effect)
        Ideally, we would like to have more demand for product
         P, but we have two choices
    1.    Generate more demand for Product P
    2.    Buy another Resource B
        Either way, we have elevated the constraint so we now
         have to execute steps 1 through 5 again.
 Systems are like chains. Every system has a weakest link, or
  constraint, which limits the success of the entire system
 Cost Accounting Systems can identify where costs are being
  incurred, but care is needed when allocating fixed costs
 TOC avoids those allocations and uses a set of three measures to
  gauge the financial impact of any decision
 Many enterprises improve performance by improving all their
  processes at the same time which removes waste but makes the
  enterprise’s focus become diffused
 Strengthening links other than the weakest link does not improve
  the strength of the chain
   The Cost World perspective is adopted by enterprises when they
     Reduce costs and trim resources so the capacity of all the resources match the
      capacity of the weakest link
     Promote local thinking and encourage behavior that runs against enterprise
   The focus should be on growth or improving throughput which is
    TOC or a Throughput World perspective
     Enterprises that adopt a Throughput World perspective have an immediate
      competitive advantage because most of the competition is stuck in the Cost
      World perspective
   The Five-Step Focusing process identifies bottlenecks and shows
    how to exploit and elevate those bottlenecks
     The Focusing process works especially with physical constraints

Shared By: